Political Entrenchment and Public Law

DARYL LEVINSON & BENJAMIN I. SACHS

Political Entrenchment and Public Law

ABSTRACT. Courts and legal scholars have long been concerned with the problem of"entrenchment" -the ways that incumbents insulate themselves and their favored policies fromthe normal processes of democratic change. But this wide swath of case law and scholarship hasfocused nearly exclusively on formal entrenchment: the legal rules governing elections, theprocesses for enacting and repealing legislation, and the methods of constitutional adoption andamendment. This Article demonstrates that political actors also entrench themselves and theirpolicies through an array of functional alternatives. By enacting substantive policies thatstrengthen political allies or weaken political opponents, by shifting the composition of thepolitical community, or by altering the structure of political decision making, political actors canachieve the same entrenching results without resorting to the kinds of formal rule changes thatraise red flags. Recognizing the continuity of formal and functional entrenchment forces us toconsider why public law condemns the former while ignoring or pardoning the latter.Appreciating the prevalence of functional entrenchment also raises a broader set of questionsabout when impediments to political change should be viewed as democratically pathologicaland how we should distinguish entrenchment from ordinary democratic politics.

AUTHORS. Daryl Levinson is the David Boies Professor of Law, NYU School of Law.Benjamin I. Sachs is the Kestnbaum Professor of Labor and Industry, Harvard Law School. Forhelpful comments and suggestions, the authors thank Will Baude, Adam Cox, Michael Farbiarz,Catherine Fisk, Heather Gerken, Jacob Gersen, Don Herzog, Sam Issacharoff, Michael Klarman,Rick Pildes, John Rappaport, Daphna Renan, Shalev Roisman, David Schleicher, David Skeel,and Adrian Vermeule, along with workshop participants at the NYU School of Law, theUniversity of Chicago Law School, and the University of Virginia School of Law. The authorsalso gratefully acknowledge the superb research assistance provided by Meredith Boak, NikolasBowie, Daniel Crossen, Alison Deich, William Dreher, Parimal Garg, Scott Hochberg, RachelHomer, Juhyung Harold Lee, Carly Rush, Mary Schnoor, Joshua Segal, and Julie Siegel.

400

ARTICLE CONTENTS

INTRODUCTION 402

1. POLITICAL ENTRENCHMENT THROUGH THE LENS OF PUBLIC LAW 408

A. What Is Political Entrenchment? 408B. Two Forms of Political Entrenchment 413

1. Electoral Entrenchment 4132. Legislative Entrenchment 4193. Common Denominators 423

II. FROM FORMAL TO FUNCTIONAL ENTRENCHMENT 426

A. The Idea of Functional Entrenchment 426B. Money and Mobilization 430

i. From Poll Taxes and White Primaries to Labor Law +322. From Campaign Finance Reform to Tort Reform 438

C. Shaping the Political Community 441

1. From Gerrymandering to State Admissions 4432. From Suffrage Restrictions to Immigration 446

D. Switching Decision Makers 448

1. From Legislative Entrenchment to Judicial Entrenchment 4492. From Legislative Entrenchment to Administrative and International

Entrenchment 451E. Summary 454

III. RATIONALIZING ENTRENCHMENT? 456

A. The Uneasy Case for Policing Formal but Not Functional Entrenchment 4571. Harmfulness 4572. Identifiability 4593. (In)conclusion 466

B. Extensions 4671. The Uncertain Case for Distinguishing Constitutional Entrenchment 4672. The Uncertain Categorical Boundaries of Entrenchment 473

CONCLUSION 477

401

THE YALE LAW JOURNAL

INTRODUCTION

In politics, winning is only the first step. The challenge is then to makevictories "stay won"-to protect them from reversal when political fortunesshift. Thus we see parties, politicians, and prevailing coalitions continuallystrategizing to lock in their gains, battening down their offices and policiesagainst the winds of political change.

As far as public law is concerned, such efforts at political "entrenchment"are viewed as dubious at best. In the context of election law, attempts bytemporarily prevailing political parties, incumbent politicians, and electoralmajorities to solidify their hold on office by gerrymandering electoral districts,selectively restricting the franchise, or using campaign finance regulation tosuppress the political speech of opponents have been the target of sustainedcriticism by scholars and some skeptical attention on the part of courts.'Manipulating the ground rules of electoral politics in these ways is regarded asan obvious pathology of democratic politics.' A separate body of scholarlycommentary and judicial decision making condemns "legislativeentrenchment" in the form of explicitly unrepealable statutes and elevatedprocedural requirements for statutory revision. Here again, the entrenchmentof political outcomes is viewed as self-evidently illegitimate: it is said to be afundamental principle of democracy that "governments are not allowed to bindfuture governments"' and that a present majority cannot "bind the hands offuture decision makers."'

Yet political actors intent on entrenching their preferred parties or policiesneed not resort to manipulating the formal rules of the Constitution, elections,or legislation. Consider recent changes to public-sector labor law. Labor unionsgenerally provide support to Democratic candidates, mobilizing pro-Democratic voters and funding the logistical and organizational infrastructureof Democratic campaigns. Seeking to defend their hold on power againstDemocratic challengers, Republican officeholders have enacted restrictive labor

1. See infra Section I.B.i.

2. See, e.g., Daniel D. Polsby & Robert D. Popper, The Third Criterion: Compactness as aProcedural Safeguard Against Partisan Gerrymandering, 9 YALE L. & POL'Y REv. 301, 305 (1991)(citing Martin Shapiro, Gerrymandering, Fairness, and the Supreme Court, 33 UCLA L. REv.227, 239 (1985)).

3. Christopher Serkin, Public Entrenchment Through Private Law: Binding Local Governments, 78U. CHI. L. REv. 879, 881 (2011).

4. Michael J. Klarman, Majoritarian Judicial Review: The Entrenchment Problem, 85 GEO. L.J.

491, 505 (1997).

402

125:4oo0 2015

POLITICAL ENTRENCHMENT AND PUBLIC LAW

legislation for the purpose of weakening unions.5 In 2011, for instance, theRepublican-dominated Wisconsin legislature overhauled the state's collectivebargaining laws to profoundly curtail unions' ability to participate effectively inpolitics. In case the purpose of these measures was not apparent, the newrestrictions exempted all the unions that had endorsed the RepublicanGovernor in the previous election.6 The goal, it seems, was to selectivelyincapacitate the Republicans' political opponents,' and not just at the statelevel: as Wisconsin's Republican senate majority leader put it at the time,"[I]f we win this battle, and the money is not there under the auspices of theunions . . . President Obama is going to have a. . . much more difficult timegetting elected . . . ." Wisconsin Republicans intent on undermining theirpolitical opposition and entrenching their party in office did not need to resortto disfranchisement or gerrymandered electoral districts. They used labor lawinstead.

Or consider Social Security, a program that is notorious for its resistance toreform or retrenchment. The program is not protected by any legal barrier torepeal or special election rules favoring its supporters. Rather, the programmobilized and empowered its defenders to stave off subsequent politicalattacks. Put differently, Social Security is entrenched not formally, but

functionally. This was no accident. In developing the program, PresidentFranklin D. Roosevelt "had one overriding aim. He wanted to entrench[S]ocial [S]ecurity so deeply in our institutional life that it would be politically

5. See Steven Greenhouse, G.O.P. Platform Seeks To Weaken Powers of Unions, N.Y. TIMES:CAucus (Aug. 30, 2012, 10:58 AM), http://thecaucus.blogs.nytimes.coM/2ol2/o8/30/g-o-p-platform-seeks-to-weaken-powers-of-unions [http://perma.cc/6ZBM-NFAJ] (describingthe 2012 Republican platform's call for elected officials to reform labor laws in restrictiveways and "salut[ing] Republican governors and state legislators" who had already takensuch steps).

6. See Governor Walker Introduces Budget Repair, OFF. GovERNoR ScoTT WALKER (Feb.11, 2011), http://walker.wi.gov/newsroom/press-release/governor-walker-introduces-budget-repair [http://perma.cc/QC64-LP8T] ("Local law enforcement and fire employees, andstate troopers and inspectors would be exempt from these changes."); Todd Richmond,Exemptions for Police, Fire Fighters in Walker Budget Bill Sparks Questions of Political Payback,Nw. (Feb. 14, 2011), http://archive.thenorthwestern.cof/article/2011o24/OSHolo1/110214045/Exemptions-police-fire-fighters-Walker-budget-bill-sparks-questions-political-payback [http://perma.cc/M8KE-95Y9].

7. See Wis. Educ. Ass'n Council v. Walker, 705 F-3d 640, 665 (7th Cit. 2013) (Hamilton, J.,concurring in the judgment in part and dissenting in part).

8. Id. at 652 (majority opinion) (second alteration in original) (quoting Wis. Educ. Ass'nCouncil v. Walker, 824 F. Supp. 2d 856, 876 n.17 (W.D. Wis. 2012)), affd in part, rev'd inpart, Walker, 705 F-3d 640.

403


impossible for his opponents to repeal it."' Or, as President Roosevelt himselfput it, "[N]o damn politician can ever scrap my [S]ocial [S]ecurity program."10

Labor law and Social Security are hardly unique. A vast literature in thesocial sciences explores the multifarious means by which political actorsinsulate themselves and their policies from political change. Examples rangewidely. In economics, Daron Acemoglu and James Robinson have argued thatthe single greatest impediment to economic growth throughout world historyhas been the conservatism of entrenched elites who fear that "creativedestruction" in the economic sphere could unsettle their dominance in thepolitical sphere." Less dramatically, in legislative contexts ranging from taxreform and emissions trading to the Affordable Care Act and Dodd-Frank,political scientists have described how progressive reformers seek to "refashionthe political context" in order to "entrench and deepen" their major policyinitiatives." Another influential body of work describes how, following the leadof New Deal Democrats who sought to build their policy gains into thestructure of the administrative state, temporarily prevailing political coalitionsseek to manipulate administrative structure and process in order to "stack thedeck" in favor of their preferred outcomes.

Legal scholars not infrequently draw upon, and even contribute to, theselines of interdisciplinary work. Yet there has been almost no recognition thatthe functional entrenchment strategies being described serve the samepurposes as the formal entrenchment techniques that public law regulates. Noris there recognition that the democratic concerns invoked against formalentrenchment are equally applicable when identical outcomes are achievedfunctionally.

Public law's normative perspective on political entrenchment is puzzling inanother respect as well. If locking in political arrangements and binding thehands of future decision makers is a democratically dubious enterprise, then

9. BRUCE ACKERMAN & ANNE ALSTOTr, THE STAKEHOLDER SOCIETY 15 (1999).

io. Paul Starr, Three Degrees of Entrenchment: Power, Policy, Structure 31-32 (Sept. 2013)(unpublished manuscript) (on file with authors).

ii. DARON ACEMOGLU & JAMES A. ROBINSON, WHY NATIONS FAIL: THE ORIGINS OF POWER,

PROSPERITY, AND POVERTY 84 (2012).

12. Eric M. Patashnik & Julian E. Zelizer, The Struggle To Remake Politics: Liberal Reform and theLimits ofPolicy Feedback in the Contemporary American State, 11 AM. POL. SCI. REV. 1071, 1072

(2013). See generally ERIC M. PATASHNIK, REFORMS AT RISK: WHAT HAPPENS AFTER MAJOR

POLICY CHANGES ARE ENACTED (2008).

13. See, e.g., Matthew D. McCubbins, Roger G. Noll & Barry R. Weingast, Structure and Process,Politics and Policy: Administrative Arrangements and the Political Control ofAgencies, 75 VA. L.

REV. 431, 433-44 (1989) (examining how legislators assure "agency compliance with thedesires of the political coalition enacting and overseeing legislation").

404

125:4oo 2015


what are we to make of constitutionalism? One of the primary purposes of theConstitution and constitutional law, after all, is to entrench rights, rules, andstructures of government against ordinary political change. To be sure, theentrenched authority of the Constitution has provoked generations ofhandwringing about the antidemocratic implications of constitutionalconstraints on present majority rule. On the whole, however, constitutionalentrenchment is widely accepted. Indeed, it is celebrated, for its contributionsto democratic stability, rights protection, and the historical continuity of theAmerican political community. What is it, then, that leads courts and scholarsto treat constitutional entrenchment as a qualitatively different phenomenonthan entrenchment at the electoral and legislative levels?

In sum, the existing picture of political entrenchment in public law is bothpartial and internally inconsistent. Courts and scholars have maintained anoddly myopic focus on entrenchment strategies that operate through explicitlegal rules aimed at processes of political change, while turning a blind, or atleast uncritical, eye to the vastly more expansive domain of politicalentrenchment. And even within that limited field of vision, public law hasregarded legislative, electoral, and constitutional entrenchment as distinct andself-contained phenomena, ignoring both their functional and normativesimilarities.

To illustrate, imagine a political coalition committed to stringent andsustained environmental regulation to prevent climate change.' Imaginefurther that the coalition has attained sufficient power at the federal level totake various kinds of political action. Finally, imagine that the coalition fearsthat its hold on power will be fleeting, and that antiregulatory political forceswill eventually regain dominance in federal politics and seek to reverse theenvironmental policies enacted by their predecessors. Here are four strategiesthe coalition might contemplate to entrench their program against repeal. Leastlikely, it could attempt to enact a constitutional amendment that guaranteescertain measures of environmental protection. Operating at the sub-constitutional level, it could attempt to enact an unrepealable environmentalstatute. Taking a less direct approach, it might instead manipulate the rules ofelection law to favor its own candidates and voters over the opposition's andtherefore retain political control and the power to continue its regulatoryagenda. Finally, it might pursue a range of functional entrenchment strategies.It could create a tradeable emissions program that would facilitate theformation of interest groups with a stake in preserving and expanding the

14. See infra notes 221-228 and accompanying text; cf Richard J. Lazarus, Super Wicked Problemsand Climate Change: Restraining the Present To Liberate the Future, 94 CORNELL L. REv. 1153(2009).

405


prevailing regulatory regime. It might try to drive polluting industries offshoreand out of the American political process. Or it could delegate expansiveregulatory authority to a politically sympathetic agency like the EnvironmentalProtection Agency, which might be more insulated from change than thepolitical branches. All of these different strategies might be viewed by thecoalition as functional substitutes -more or less interchangeable mechanismsfor accomplishing the same basic purpose. But public law would view them asquite distinct, as a matter of both legal rules and normative democratic theory.

This Article questions what, if anything, justifies this differential treatment.At a descriptive level, it catalogues and compares the range of legal and politicaltechniques through which parties, politicians, and policies are insulated againstcontestation and change. At a normative level, it raises questions aboutwhether and when political entrenchment of various kinds should be regardedas a matter of concern in public law and what exactly the concern should be.

More specifically, the Article proceeds as follows. Part I surveys how thephenomenon of political entrenchment has been defined and regulated as amatter of public law. Entrenchment comes into view when political actorsintentionally create legal impediments to political change. Beyond the specialcase of constitutionalism, public law has recognized and regulated thisbehavior primarily in two contexts.

One is election law, where scholars have increasingly viewed theentrenchment of incumbent officeholders, political parties, and majoritycoalitions as the central problem that legal regulation of the political processshould be designed to solve. Although courts have not yet fashioned doctrinaltools aimed explicitly at preventing or remedying entrenchment, judges andJustices have joined in the scholarly skepticism and in some cases have foundways of striking down election rules that seemed to have the purpose and effectof suppressing democratic competition and protecting power holders againstpolitical challenge. The doctrinal prohibition on entrenchment is more explicitin the second context of legislative entrenchment. It has long been understoodthat legislatures are not permitted to enact unrepealable statutes or to insulatestatutes against repeal or revision by way of supermajority rules or other specialprocedural requirements. The blurry boundaries of this prohibition have beeninterpreted inconsistently by judges and scholars, who have invoked it to castdoubt on a whole range of laws, from government contracts to frameworkstatutes and the Senate filibuster.

Courts and scholars have understood electoral and legislative entrenchmentas separate and independent phenomena, but it may be more illuminating toview them as pieces of a larger puzzle. Political actors use electoralentrenchment to accomplish indirectly what legislative entrenchmentaccomplishes directly, namely, insulating substantive policy outcomes againstshifting political preferences.

4o6

125:4oo0 2015


Electoral and legislative entrenchment (as well as constitutionalentrenchment) are created by means of formal legal rules governing processesof political change-the rules governing voting and elections, the enactment orrepeal of legislation, and constitutional adoption and amendment. Yet, as PartII describes, politicians, parties, and policies can be entrenched throughfunctional, political mechanisms just as readily as through formal, legal ones.Developing and drawing upon a wide range of examples, this Part synthesizesthree general mechanisms of functional entrenchment. First, politicians,parties, and temporarily prevailing coalitions can enact substantive policies thatstrengthen political allies or weaken political opponents. Second, they canenact policies or programs that change the composition of the politicalcommunity, selecting in allies or selecting out opponents. Third, they can shiftthe locus of political decision making to an actor or institution that isresponsive to allies or unresponsive to opponents. These functional strategiesappear to be close substitutes for formal electoral, legislative, and constitutionalentrenchment, and there is every reason to believe they are widely used bypolitical actors to accomplish the same ends.

Why does public law view formal entrenchment as a form of democraticfailure and an attractive target for legal regulation while treating functionalentrenchment largely as a matter of normative and legal indifference? Part IIItakes up this question, considering whether the apparent inconsistency can beexplained or rationalized. Perhaps formal entrenchment is more harmful todemocratic values, less susceptible to benign or beneficial uses, or simply easierto identify and police? Section III.A considers these possibilities but finds themless than fully persuasive. The remainder of Part III goes further in a skepticaldirection. Section III.B.1 asks whether there is any good reason for viewingconstitutional entrenchment more favorably than legislative or electoralentrenchment, or even for treating it as a different category. Section III.B.2raises the question of whether, once we recognize that political entrenchment isnot limited to formal entrenchment, the concept has any clear outer boundariesor coherent core. A unifying theme of the discussion in Part III -amplified inthe Conclusion-is the need for a broader perspective on impediments topolitical change and assessments of their costs and benefits or democraticlegitimacy.

The progression of the argument along these lines leads to a shift inperspective that it may be helpful to foreshadow. Our main thrust is to showthat the formal arrangements commonly described by the term "entrenchment"have functional analogues that are driven by the same motivations and havesimilar effects. For this purpose, we proceed on the assumption that "politicalentrenchment" is an adequately -even if not always clearly or consistently -defined phenomenon. Our argument is that on any plausible understanding ofentrenchment, there will be innumerable political phenomena that fit the bill,

407


beyond the narrow band of formal entrenchment. Once we push past theformal markers of electoral and legislative entrenchment to focus on thefunctional goals and mechanisms that might be viewed as entrenching,however, the boundaries of the category of "political entrenchment" and thefeatures that are supposed to distinguish it from ordinary or desirabledemocratic politics begin to fade. By appreciating the potential breadth of thecategory of entrenchment, this Article not only expands our understanding ofthat phenomenon but also ultimately calls into question its meaning andutility.

1. POLITICAL ENTRENCHMENT THROUGH THE LENS OF PUBLIC LAW

A. What Is Political Entrenchment?

Political "entrenchment" is discussed more often than it is defined, and it isnot clear that any single definition captures all uses of the term. At the mostgeneral level, "entrenchment" means that political change has been made moredifficult than it otherwise would (or should) be."s "Political change" isobviously a broad category. At the level of constitutionalism, the relevantobjects of stasis and change include the structure of government, theboundaries and allocation of governmental powers, and the set of rules andrights prohibiting specific governmental actions. At the subconstitutional level,political change can mean change in which politicians or parties are elected tooffice or change in the substantive policy outcomes generated by these powerholders and their supporters.

Impediments to political change can take a number of different forms.Public lawyers tend to focus on formal, procedural barriers to change, such asthe Article V requirements of dual supermajorities for constitutionalamendment or a hypothetical statute that deems itself unrepealable. The legalrules governing political change through the democratic process are also acommon target of entrenchment concerns. Parties that disfranchise or suppressthe political speech of opponents, incumbent legislators who gerrymanderelectoral districts to ensure their own reelection, and dictators who outlawopposition parties or cancel elections altogether are all engaged in projects ofpolitical entrenchment, manipulating the ground rules of the democraticprocess in order to retain their hold on power.

408

15. Cf Starr, supra note lo, at 2 (defining entrenchment as "the creation of mechanisms toimpede or constrain ordinary or expected processes of change" (emphasis omitted)).

125:4oo 2015


As we shall emphasize, however, manipulating formal rules is not the onlyway to prevent change. After all, dictators can imprison or shoot theiropponents rather than disfranchise them. Less dramatically, parties, politicians,and policies can create political, rather than legal, impediments to change.Recall the introductory example of labor law reform: incumbents can entrenchthemselves in office not only through gerrymandering or franchise restrictionsbut also by incapacitating the electoral organization of the political opposition.Or recall the example of Social Security: the program is difficult to retrench notbecause of any legal barrier to repeal, but because the enactment of theprogram mobilized and empowered defenders to stave off subsequent politicalattacks.'

Whatever form impediments to political change might take, to qualify as"impediments" they must be distinguishable from the expected workings ofthe political process. Political entrenchment implies not just the absence ofpolitical change but some kind of special constraint on the usual processes ofpolitical change. Thus, the persistence of politicians or parties in office, or thepreservation of particular policies over long periods of time, is not necessarilyproof of entrenchment. If politicians, parties, or policies are retained simplybecause they continue to be popular among the electorate, this would not beviewed as entrenchment. Entrenchment implies that the political system is notresponsive to changes in voters' preferences; a system that is perfectlyresponsive to unchanging preferences would be viewed as a well-functioningdemocracy.7

Thus, notwithstanding conventional claims to the contrary, it is possiblethat Social Security has proven politically durable simply because politicalsupport for the goal of providing financial security for people in old age has notdiminished over the past eighty years. If this were the complete explanation forthe program's survival, we should not think of Social Security as entrenchedany more than we think of criminal laws against homicide as entrenched."Both might endure simply because they remain consistent with the first-orderpolitical preferences of a (super)majority of citizens. The perception that SocialSecurity is entrenched stems from the view that, in contrast to prohibitions onmurder, a present majority might not vote to reenact the program in anythinglike its current form. The program persists, in this view, because it is nowdefended by a powerful interest group, brought into being by the program

i6. See supra notes 9-1o and accompanying text.

17. See Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of ConstitutionalCommitment, 124 HARv. L. REv. 657, 702 (2011); see also Starr, supra note lo, at 1("Entrenchment is not the same as persistence, though it can be one of its causes . . .

18. Levinson, supra note 17, at 702.

409


itself, which has proven capable of preventing present majority preferencesfrom prevailing.

Other kinds of impediments to political change blur the boundary betweenentrenchment and ordinary politics. Suppose that Social Security persists not(just) because of interest group mobilization but because of its increasingpopularity over time, as Americans have learned from their experience underthe program that mandatory savings for retirement is more beneficial than theyinitially imagined. One could view this dynamic of endogenous preferencechange as a mechanism of entrenchment on the theory that this kind of path-dependent increase in political support should count as a special impediment toordinary political change. Or suppose that critical support for preserving SocialSecurity stems from the expectation among workers that the earmarked taxesthey have paid into the program are now owed to them by the governmentupon retirement, or by the reliance of many Americans on the existence ofSocial Security payments to support their retirement, leading them not to savethrough other vehicles." One could also view these kinds of adaptivepreference shifts as mechanisms of entrenchment.

For the purposes of this Article, however, we will work with a more limiteddefinition of entrenchment. Rather than regarding some kinds of shifts inpreferences as creating entrenchment barriers, we shall take individual politicalpreferences, regardless of how they have been shaped or transformed, as given.Furthermore, we shall accept the satisfaction of present majority will -again,"black boxed" with respect to the process of its formation - as a benchmark forwell-functioning democracy. Only impediments to giving effect to presentmajority will, such as supermajority rules for revising statutes or politicaldynamics like the mobilization of a powerful interest group, will be taken asexamples of entrenchment.20

Identifying this kind of entrenchment requires some baseline conception ofordinary, unconstrained processes of political change. In the public lawliterature on entrenchment, two kinds of "ordinary politics" baselines arecommonly in play.'

One is the process for, or political difficulty of, effecting change throughsome alternative channel, usually one that is more responsive to majority will. '

ig. On these alternative explanations for the political entrenchment of Social Security, see Starr,supra note io, at 31-33.

20. See infra Section III.B.2 for a discussion of this limited definition of entrenchment.

21. See infra Section III.B.2, which revisits these definitional baselines and questions theirutility in demarcating a limited category of entrenchment. For now, we are attempting aworking definition.

22. See Klarman, supra note 4, at 498 (conceptualizing entrenchment as antimajoritarian).

410

125:40o 2015


For example, when constitutional law in the United States is described asentrenched, reference is typically made to the "supermajorities" needed toeffect constitutional change, in contrast to the "majorities" needed to enact astatute. This is obviously a highly stylized, even formalistic, vision of howactual lawmaking processes operate. With respect to statutes, we might pushpast the caricature of "majority rule" to notice, for example, the differentmajorities implicated by electing senators, representatives, and the President,and the likely supermajorities necessary to assemble a prevailing legislativecoalition. Moreover, the procedural barriers to statutory enactment would beonly part of a functional assessment of political difficulty, which would dependon many other variables - the formation and alignment of coalitions, the abilityof interest groups to block action, internal legislative procedures and agenda-setting power, and much else." For present purposes, however, the importantthing to see is that the baseline being used to define entrenchment is the (moreor less hypothetical) alternative of effecting political change through someprocess that (better) tracks the preferences of democratic majorities or themedian voter.

A different "ordinary politics" baseline is set by the degree of difficulty ofcreating the status quo. Under this standard, entrenchment means that a politicalarrangement is now more difficult to change than it was to create in the firstplace.' On this definition, it is no longer clear that the U.S. Constitutionshould count as entrenched, because it is not obvious that thesupermajoritarian procedures required for amendment are a higher hurdle thanthe supermajoritarian procedures the Constitution had to overcome in thecourse of its initial enactment." Likewise, an unpopular incumbent who cannotbe dislodged from office is not entrenched if she originally had to defeat asimilarly advantaged predecessor to win her post.

In many cases the two criteria for identifying entrenchment converge. Inthe paradigmatic case of legislative entrenchment, for example, a statute

23. For a discussion of these complexities and their consequences for the way we might thinkabout entrenchment, see infra Section III.B.2.

24. See Eric A. Posner & Adrian Vermeule, Legislative Entrenchment: A Reappraisal, iii YALE L.J.

1665, 1667 (2002) (defining legislative entrenchment as "the enactment of either statutes orinternal legislative rules that are binding against subsequent legislative action in the same

form" (emphasis added)); see also John 0. McGinnis & Michael B. Rappaport, SymmetricEntrenchment: A Constitutional and Normative Theory, 89 VA. L. REv. 385, 388-89 (2003)

(distinguishing between "symmetric" and "asymmetric" entrenchment).

25. The unamendable requirement that no state be denied equal suffrage in the Senate is anexception. See U.S. CONsT. art. V. That provision is procedurally more difficult to repealthan it was to enact. See McGinnis & Rappaport, supra note 24, at 411-15; Posner &Vermeule, supra note 24, at 1681-82.

411


enacted by majority vote specifies that a supermajority is required for revisionor repeal, combining an upward departure from the absolute standard ofmajority rule with an upward departure from the relative standard for initialenactment. Likewise, in the paradigmatic case of electoral entrenchment, aparty or coalition manipulates the rules of election law upon gaining office-forinstance, by disfranchising opponents or reducing their voting power-suchthat a subsequent majority of voters who would prefer to replace theincumbents will be thwarted. If that same majority would have been sufficientto prevent the incumbents from being elected in the first place, then bothcriteria of entrenchment are satisfied. Most of the examples this Articlediscusses qualify as entrenched according to both baselines. Social Security, forinstance, might be classified as entrenched both by reference to presentmajority will and by reference to the initial difficulty of the program'senactment, prior to the formation of a mobilized group of vested beneficiariesand supporters.

Public law has primarily focused on entrenchment as an intentionalstrategy,26 and most of the examples we discuss are of this sort.' Theintentionality of entrenchment is often associated with bad motives, as whenparties and politicians engage in self-serving efforts to suppress competitionand maintain their hold on power. But intentional entrenchment need not beself-serving. As the literature on constitutionalism emphasizes, there areperfectly respectable, public-regarding reasons for entrenchment.Constitutions, in common with other mechanisms of entrenchment, facilitateenduring political commitments (or "precommitments"), protectingnormatively preferred policies from being undermined by shortsighted orotherwise pathological decision making." Constitutional and other forms ofentrenchment also promote political coordination and stability, reducing thecosts of both conflict and transition.

Criticisms of entrenchment are also familiar from the literature onconstitutionalism. Entrenched policies and political arrangements arguablysubstitute rule by the "dead hand" of the past for rule by present majorities,

26. See, e.g., Samuel Issacharoff & Richard H. Pildes, Politics as Markets: Partisan Lockups of theDemocratic Process, 50 STAN. L. REV. 643, 644 (1998) (exploring "ways in which dominantparties manage to lock up political institutions to forestall competition"); Klarman, supranote 4, at 502 (starting from the assumption that "legislators strongly prefer to remain inoffice").

27. Section III.B.2 revisits intentionality as a criterion of entrenchment.

28. See JON ELSTER, ULYSSES UNBOUND: STUDIES IN RATIONALITY, PRECOMMITMENT, AND

CONSTRAINTS 88-174 (2000); STEPHEN HOLMEs, PASSIONS AND CONSTRAINT 134-77 (1995).

412

125:40 0 2015


threatening democratic ideals of popular sovereignty and self-government.'The dubious democratic legitimacy of entrenchment goes hand in hand withpractical concerns about preventing the current political community fromresponding to changed circumstances or shifting values by locking in bad oranachronistic policy decisions. We return to the costs and benefits of politicalentrenchment below."o

B. Two Forms ofPolitical Entrenchment

Bracketing the special case of constitutionalism," public law has grappledmost extensively with political entrenchment in two contexts. Electoralentrenchment involves efforts by parties and politicians to entrench themselvesin office by manipulating the rules of democratic politics. Such efforts havebeen generally frowned upon by courts and commentators, and scholars havecalled for broad swathes of election law jurisprudence to be reoriented towardpreventing political entrenchment of this kind. Moving from elections togovernance, it has long been assumed that "legislative" entrenchment-including, at a minimum, the enactment of statutes that cannot be revised orrepealed by a majority of a subsequent legislature-is constitutionallyimpermissible and democratically illegitimate.

1. Electoral Entrenchment

In democratic politics, power holders-whether incumbents, politicalparties, or electoral coalitions-will often possess the means and motivation topreserve their privileged positions by rigging the rules of the electoral system.In some cases, the desire of elected officials to entrench themselves in officemay lead them to act contrary to the preferences of their constituents. Thus,term limits have found little support among incumbent state legislators, whopredictably lack enthusiasm for voting themselves out of a job., In other cases,officeholders and their constituents will share a common interest inperpetuating their hold on power and in fending off political challenges from

2g. See DAVID A. STRAuss, THE LIVING CONSTITUTION 99-102 (2010).

30. See infra Part III.

31. We return to constitutionalism infra Section III.B.i.

32. See Klarman, supra note 4, at 509-13. Even where clear majorities or supermajorities ofvoters support term limits, in most jurisdictions the only route to their enactment has beenthrough initiative and referendum processes that bypass legislatures. See id. at 510.

413


opposing parties or coalitions.' Both types of electoral entrenchment areviewed as predictable pathologies of the democratic political process andproblems that law might help to solve.

Electoral entrenchment strategies take many different forms. The moststraightforward is simply to prevent one's opponents or their supporters fromcasting ballots, while enfranchising as many of one's own supporters aspossible. Thus, after the Civil War, Republicans in Congress sought toenfranchise black voters in the South, partly for moral and ideological reasons,but also to ensure the electoral dominance of their party.' The end ofReconstruction allowed Southern Democrats to turn the tables, using force,fraud, poll taxes, literacy tests, and other tactics to disfranchise virtually allblack voters and many poor whites, thereby restoring and entrenching theirown political supremacy." Through the 196os (and perhaps beyond), blackdisfranchisement was used as a tool of entrenchment for the Democratic partyin the South; factions within the party; elected officials who might bevulnerable to defeat by black voters (or biracial coalitions); and, of course,white majorities, which were able to maintain political and social dominance bymonopolizing control over government. In recent elections, voter ID laws,more stringent registration requirements, the curtailment of early voting, andother procedural regulations have been supported or opposed based in largepart on their predictable effects on the racial and partisan composition of theelectorate: Democrats accuse Republicans of supporting voter ID and similarprocedural requirements in order to disproportionately exclude Democraticvoters, while Republicans accuse Democrats of opposing voter ID laws so thatmore illegal ballots will be cast in favor of Democrats.

33. See id. at 498 (distinguishing the former kind of "agency" problem from the problem of"cross-temporal majorities," while portraying both as problematic forms of electoralentrenchment).

34. See ALEXANDER KEYSSAR, THE RIGHT To VOTE 87-104 (2000); MICHAEL J. KLARMAN, FROM

JIM CROW TO CIVIL RIGHTS 28-29 (2004).

3s. See KEYSSAR, supra note 34, at 105-16. See generally J. MORGAN KOUSSER, THE SHAPING OF

SOUTHERN POLITICS: SUFFRAGE RESTRICTIONS AND THE ESTABLISHMENT OF THE ONE-PARTY

SOUTH, 1880-1910 (1974).

36. See Issacharoff& Pildes, supra note 26, at 660-66 (describing how white conservative elitesin Texas and elsewhere in the South used black disfranchisement to entrench their controlover the Democratic Party and the Party's control over the state).

37. See Nicholas 0. Stephanopoulos, Elections and Alignment, 114 COLUM. L. REV. 283, 324-30(2014) (describing the new array of franchise restrictions and their partisan

consequences); see also Maggie Haberman & Amy Chozick, Democrats Wage a National Fightover Voter Rules, N.Y. TIMES (June 3, 2015), http://www.nytimes.com/2o15/o6/o4/us/politics/democrats-voter-rights-lawsuit-hillary-clinton.html [http://perma.cc/55HQ-BMCL] (describing the burgeoning legal battles over voter access restrictions and

414

125:4oo0 2015


Another time-honored technique for tilting the electoral playing field ismanipulating the number, size, and boundaries of electoral districts-theinfamous gerrymander. Before the Supreme Court required equipopulosity oflegislative districts, legislators elected from malapportioned districts resistedany change in district boundaries, just as constituents in overrepresenteddistricts resisted reapportionment schemes that would reduce theirrepresentation." Along with at-large and multimember districting schemes,gerrymandering was a key line-drawing tool used by white majorities,incumbent legislators, and the Democratic Party in the South to suppress blackvoting power and preserve political hegemony.39 Partisan gerrymanders remaina staple of contemporary politics, permitting parties to leverage temporary orslight legislative majorities into enduring or decisive control without thetrouble of attracting more votes.40 Alternatively, legislators who manage toovercome their partisan differences and cooperate across party lines have theopportunity to agree on districting schemes designed to preserve the safety oftheir seats - so-called "bipartisan" or "incumbent" gerrymanders.4 '

Many other levers of electoral entrenchment are available to strategicpolitical actors. Political parties that gain effective control of government canregulate the party structure of elections and have done so with predictableattention to the prospects for their own electoral success - for example, byrequiring closed primaries when their competitor party would benefit from anopen primary structure.' Or, the two major parties can collaborate to protecttheir "duopoly" by using cumbersome ballot access requirements,' bans on

their partisan stakes in the 2016 presidential election cycle); Richard L.Hasen, The Voting Wars Heat Up, SLATE (Sept. 29, 2014), http://www.slate.com/articles/news and-politics/jurisprudence/2o14/o9/voting-restrictions may_reach-the-supreme-court fromohio wisconsinnorth.html [http://perma.cc/6FzF-9FXM] (surveying litigation over restrictive voting legislation and stressing the partisanstakes).

38. See Klarman, supra note 4, at 513-15.

3g. See Issacharoff & Pildes, supra note 26, at 700-03.

40. See Stephanopoulos, supra note 37, at 348-49 (presenting evidence that, in recent decades,parties with full control over state governments have enacted districting plans that awardthemselves six percent more seats on average than the plan that would have resulted if theopposing party had been in charge of redistricting); see also id. at 286 (suggesting thatgerrymandering helped Republicans keep their majority in the House in 2012 despitereceiving 1.4 million fewer votes nationwide than Democrats).


42. See Richard H. Pildes, Foreword: The Constitutionalization of Democratic Politics, 118 HARv. L.REv. 28, 102 n.298 (2004).


4.15


fusion candidacies,' or "sore loser" laws to prevent entry by third parties orindependent candidates.4s Campaign finance regulation offers yet anothertempting instrument for suppressing competition and securing political power,allowing incumbents or temporarily dominant parties to channel money tothemselves and away from challengers, while also helping corporations andwealthy donors protect their preferred policies against challenges from lesswealthy constituencies .46

Courts have intervened in all of these areas, developing an elaboratejurisprudence governing many facets of the electoral process. Poll taxes, literacytests, and other instruments of minority disfranchisement have beeninvalidated.47 The constitutional rule of "one person, one vote" now governsthe drawing of electoral districts.8 Race-conscious gerrymandering ismandated by the Voting Rights Act to ensure a measure of minorityrepresentation, but also constrained by the Equal Protection Clause to avoidoverly or too overtly race-based decision making.49 The Supreme Court hasdeemed political gerrymandering a constitutionally cognizable problem, albeitone for which the Justices have not been able to agree upon a judiciallymanageable solution.so Most limitations on campaign spending, beyond theregulation of direct contributions to candidates, have been invalidated as

44. See Issacharoff& Pildes, supra note 26, at 683-86; Pildes, supra note 42, at 117-26.

45. See Michael S. Kang, Sore Loser Laws and Democratic Contestation, 99 GEO. L.J. 1013, 1042-58(2011).

46. See Klarman, supra note 4, at 522-23; Pildes, supra note 42, at 130-53.

47. See, e.g., Thornburg v. Gingles, 478 U.S. 30, 80 (1986) (invalidating a multimemberdistricting scheme for its discriminatory effect on black voters); Oregon v. Mitchell, 400U.S. 112, 131-34 (1970) (reaffirming Congress's ban on literacy tests as a validantidiscrimination measure); Katzenbach v. Morgan, 384 U.S. 641, 646-47 (1966)(upholding sections of the Voting Rights Act that restricted literacy tests for certain non-English speaking citizens); Harper v. Va. Bd. of Elections, 383 U.S. 663, 666 (1966)(holding that Virginia's poll tax violated the equal protection clause of the fourteenthamendment).

48. See, e.g., Reynolds v. Sims, 377 U.S. 533, 558-71 (1964); Wesberry v. Sanders, 376 U.S. 1, 17-18 (1964).

49. See, e.g., Shaw v. Reno, So9 U.S. 630 (1993) (holding that redistricting based on race isevaluated under strict scrutiny, yet requiring redistricting to be race-conscious to ensurecompliance with the Voting Rights Act).

so. See Vieth v. Jubelirer, 541 U.S. 267 (2004) (plurality opinion) (deciding, in a split decisionwith no majority opinion, not to intervene in a congressional redistricting plan); Davis v.Bandemer, 478 U.S. 1o9 (1986) (holding that political gerrymandering claims arejusticiable, but without a majority agreement upon a standard to govern such claims).

416

125:40oo 2015


violations of free speech." Some types of regulations of political parties andballot access limitations have also been rejected as unconstitutional.52

While many of these judicial interventions have had the effect of limitingopportunities for political entrenchment, entrenchment has not typically beenthe doctrinal focus. Instead, courts have tended to frame their role as enforcingindividual rights, leaving systemic concerns like preserving politicalcompetition and preventing entrenchment mostly offstage.5 Nonetheless,recognition and disapproval of electoral entrenchment not infrequently bubbleto the surface of judicial opinions. Motivating the Court's initial decision toenter the "political thicket" was the recognition that malapportionmentthreatened "systematic frustration of the will of a majority of the electorate,"4

and that "entrenched political regimes" prevented a legislative solution.55

Courts have been especially skeptical of ballot access restrictions imposed uponthird parties and independent candidates when these restrictions "operate tofreeze the political status quo."56 In a Seventh Circuit decision ultimatelyaffirmed by the Supreme Court, an Indiana voter ID law was upheld over adissent that described the law as "a not-too-thinly-veiled attempt to discourageelection-day turnout by certain folk believed to skew Democratic."' Expressing

s. See, e.g., McCutcheon v. FEC, 134 S. Ct. 1434 (2014); Am. Tradition P'ship, Inc., v. Bullock,

132 S. Ct. 2490 (2012) (per curiam); Citizens United v. FEC, 558 U.S. 310 (2010).

52. See, e.g., Anderson v. Celebrezze, 460 U.S. 780 (1983) (declaring Ohio's filing deadline forindependent candidates unconstitutional); Williams v. Rhodes, 393 U.S. 23 (1968) (holdingthat Ohio's restrictive ballot access laws violated the equal protection clause by effectivelylimiting access to the two major parties).

53. See Issacharoff& Pildes, supra note 26, at 644-46, 717; Pildes, supra note 42, at 40-41.

54. Lucas v. Forty-Fourth Gen. Assembly, 377 U.S. 713, 753-54 (1964) (Stewart, J., dissenting);see also Klarman, supra note 4, at 531-32.

ss. Baker v. Carr, 369 U.S. 186, 248 (1962) (Douglas, J., concurring); see also Reynolds v. Sims,377 U.S. 533, 553-54, 570 (1964) (noting that the entrenchment of the Alabama legislatureprevented the development of a solution to alleged malapportionment); Klarman, supranote 4, at 531.

56. Jenness v. Fortson, 403 U.S. 431, 438 (1971); see also Klarman, supra note 4, at 535-36. TheSupreme Court has invalidated a closed primary requirement imposed on the minorityRepublican Party by the Democratic-controlled legislature, i.e., "the one political partytransiently enjoying majority power." Tashjian v. Republican Party of Conn., 479 U.S. zo8,224 (1986).

57. Crawford v. Marion Cty. Election Bd., 472 F.3d 949, 954 (7th Cir. 2007) (Evans, J.,dissenting), affd, 553 U.S. 181 (2008). Judge Posner, who authored the majority opinion for

the Seventh Circuit panel, subsequently came to share the dissenting view. Referencing the"ferocity" of party competition, Judge Posner admitted in an interview that he "wasn't alertto this kind of trickery, even though it's age old in the democratic process." John Schwartz,Judge in Landmark Case Disavows Support for Voter ID, N.Y. TIMEs (Oct. 15, 2013),

417


skepticism of campaign finance regulation, Justice Scalia has warned that"[t]he first instinct of power is the retention of power, and, under aConstitution that requires periodic elections, that is best achieved by thesuppression of election-time speech."s8 And Justice Breyer has explained theconstitutional problem with partisan gerrymanders in terms of "[t]hedemocratic harm of unjustified entrenchment," evidenced by a redistrictingplan that awards a party receiving a minority of statewide votes a majority oflegislative seats."

Prominent election law scholars have been more overtly and consistentlyfocused on entrenchment as a central concern for legal regulation of thepolitical process. For example, Michael Klarman argues that courts shouldcommit themselves to policing the dual entrenchment problems ofrepresentatives perpetuating their hold on office by acting contrary to thewishes of their constituents and temporary political majorities seeking toextend their hold on power into the future.6 o To this end, Klarman develops aframework for "anti-entrenchment review" of districting, disfranchisement,ballot access restrictions, campaign finance reform, and other areas of electionlaw.6' Klarman's approach is motivated by an overarching commitment to thedemocratic value of majority rule, which he sees as threatened wheneverofficials contradict the preferences of a majority of citizens or when the will of apresent majority is thwarted by entrenched arrangements.2 Similarly focusedon the problem of entrenchment, Samuel Issacharoff and Richard Pildesemphasize the need to maintain political competition and to guard against"political lockups" perpetrated by "existing holders of political power [who]seek to perpetuate their control . . . by capturing the basic structures andground rules of politics itself."6

' As Pildes elaborates, judicial intervention isjustified "whenever self-interested political actors employ political power toinsulate themselves from the political competition required to make electoralaccountability meaningful."64

http://www.nytimes.com/2o13/lo/16/us/politics/judge-in-landmark-case-disavows-support-for-voter-id.html [http://perma.cc/G3T9-AWMK].

58. McConnell v. FEC, 540 U.S. 93, 263 (2003) (Scalia, J., concurring in part, concurring in thejudgment in part, and dissenting in part).

59. Vieth v. Jubelirer, 541 U.S. 267, 361 (2004) (Breyer, J., dissenting).

6o. See Klarman, supra note 4, at 498.

61. See id. at 528-39.

62. See id. at 502-09.

63. Issacharoff& Pildes, supra note 26, at 648, 650.64. Pildes, supra note 42, at 46.

418

125:40 0 2015


2. Legislative Entrenchment

It has long been conventional wisdom among constitutional lawyers that"one legislature may not bind the legislative authority of its successors."s6 Morespecifically, a legislature may not "entrench" a law by forbidding subsequentrepeal or amendment, or by imposing heightened procedural hurdles, such assupermajority voting rules that were not necessary to enact the law in the firstplace.66 For example, Congress would not be permitted to enact a statuterequiring a balanced federal budget "in perpetuity," or with an attachedprohibition on repeal, or a prohibition on repeal by less than a two-thirdsmajority. If Congress did enact such a statute, the purported entrenchmentwould presumably be invalidated by courts (to the extent they would find theissue justiciable). And it could be legally ignored by subsequent Congresses:notwithstanding the statutory language, a congressional majority in pursuit ofan unbalanced budget would be free to repeal or override the preexistingstatute pursuant to the standard second-in-time rule. This, at least, is theconsensus view among constitutional theorists.

The precise source of the anti-entrenchment principle in U.S.constitutional law has never been entirely clear. Aversion to legislativeentrenchment has a long history in British constitutional thought, where-atleast in theory, if not always in practice68 - " [t]here is no law which Parliamentcannot change"69 and "[a]cts of parliament derogatory from the power ofsubsequent parliaments bind not."'o But the British version of the anti-entrenchment principle developed as a corollary of parliamentary supremacy,and so it does not obviously translate to the American legal system, in which

65. United States v. Winstar Corp., 518 U.S. 839, 872 (1996) (plurality opinion) (referencing iWILLIAM BLACKSTONE, COMMENTARIES *90).

66. Thus, Eric Posner and Adrian Vermeule define legislative entrenchment as "the enactmentof either statutes or internal legislative rules that are binding against subsequent legislativeaction in the same form." Posner & Vermeule, supra note 24, at 1667.

67. See, e.g., LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW 124-25 na- (3d ed. 2ooo);

David Dana & Susan P. Koniak, Bargaining in the Shadow of Democracy, 148 U. PA. L. REv.473, 526-36 (1999); Julian N. Eule, Temporal Limits on the Legislative Mandate: Entrenchmentand Retroactivity, 1987 AM. B. FOUND. RES. J. 379; Paul W. Kahn, Gramm-Rudman and theCapacity of Congress To Control the Future, 13 HASTINGS CONST. L.Q 185 (1986). But seePosner & Vermeule, supra note 24 (arguing that there is no basis for a rule against legislativeentrenchment).

68. See Posner & Vermeule, supra note 24, at 1678.

69. A.V. DICEY, INTRODUCTION TO THE STUDY OF THE LAw OF THE CONSTITUTION 84 (8th ed.1915).

70. 1 WHLIAM BLACKSTONE, COMMENTARIES *90.

419


the legislature is subordinate to the Constitution.' In its American incarnation,the prohibition on legislative entrenchment seems to have been recast as ameans of marking the contrast between entrenched constitutional law andordinary law lacking this definitive constitutional characteristic. As LaurenceTribe once testified to Congress, "Only by a constitutional amendment can onetruly bind the future: unless we keep clearly in mind that distinction between aconstitutional amendment and a bill or resolution, we have really lost ourway."7 2

As a textual matter, the anti-entrenchment principle has been variouslygrounded in some combination of the Article I grant of limited legislativepowers, the provisions of Article I specifying limited terms of office forcongressional representatives, and Article V, which has been understood ascreating an exclusive pathway for supra-statutory entrenchment.3 TheSupreme Court has enforced the rule in a handful of cases, though withoutmuch explication of its constitutional source, justification, or scope. Forinstance, in holding that the Ohio State Legislature was free to change thelocation of a county seat notwithstanding a preexisting statute that had"permanently established" the existing seat,' the Court explained:

Every succeeding legislature possesses the same jurisdiction and powerwith respect to [the public interest] as its predecessors. The latter havethe same power of repeal and modification which the former had ofenactment, neither more nor less. All occupy, in this respect, a footingof perfect equality. This must necessarily be so in the nature of things.It is vital to the public welfare that each one should be able at all timesto do whatever the varying circumstances and present exigenciestouching the subject involved may require. A different result would befraught with evil."

Although the prohibition on statutory entrenchment has gone largelyuntested -neither Congress nor state and local legislatures appear to have

71. See United States v. Winstar Corp., 518 U.S. 839, 872 (1996) (plurality opinion); Eule, supranote 67, at 393.

72. Extending the Ratification Period for the Proposed Equal Rights Amendment: Hearings on H.R.J.Res. 638 Before the Subcomm. on Civil and Constitutional Rights of the H. Comm. on theJudiciary, 95th Cong. 51 (1977) [hereinafter Hearings] (statement of Laurence H. Tribe,Professor of Law, Harvard Law School).

73. See Posner & Vermeule, supra note 24, at 1680-95 (critically surveying these and othertextually grounded arguments).

74. Newton v. Comm'rs, 1oo U.S. 548, 561 (1879).

7s. Id. at 559.

420

125:4oo 2015


attempted anything like this very often76 - the anti-entrenchment principle hasbeen extended to analogous legislative acts that are more prevalent in the realworld. For example, the principle has been invoked to criticize theentrenchment of the Senate's cloture rule requiring sixty votes to end afilibuster, and to argue that, as a constitutional matter, a simple majority mustbe empowered to end filibusters.n The principle has also been cited inobjections to "framework" statutes like the Gramm-Rudman Act, theAdministrative Procedure Act, and the War Powers Resolution, which seek toimpose constraints upon, or take presumptive priority over, downstreamlegislative decision making. Scholars have also raised anti-entrenchmentobjections to the creation of property rights protected against subsequentconfiscation by compensation requirements" and to consent decrees that lock

80in government policies against subsequent revision.

As far as courts have been concerned, the anti-entrenchment principle hashad the most purchase in the constitutional law of government contracting.Judicial enforcement of contracts entered into by earlier legislatures againsttheir successors pursuant to the Contracts and Takings Clauses lookssuspiciously similar to legislative entrenchment.8

, Since the early Republic, the

76. See Posner & Vermeule, supra note 24, at 1678-79.

77. See Eule, supra note 67, at 407-15; Catherine Fisk & Erwin Chemerinsky, The Filibuster, 49STAN. L. REv. 181, 250 (1997); Posner & Vermeule, supra note 24, at 1694-95; John C.Roberts & Erwin Chemerinsky, Entrenchment of Ordinary Legislation: A Reply to ProfessorsPosner and Vermeule, 91 CAUF. L.REV. 1773, 1780-81 (2003).

78. Cf Amandeep S. Grewal, Legislative Entrenchment Rules in the Tax Law, 62 ADMIN. L. REv.1011 (2010) (discussing the Administrative Procedure Act and the War Powers Resolution);Kahn, supra note 67 (discussing the Gramm-Rudman Act); Posner & Vermeule, supra note24, at 1695-97 (discussing the Gramm-Rudman Act). Statutes that purport to create rules ofstatutory interpretation or construction for downstream statutes, such as the ReligiousFreedom Restoration Act or the Defense of Marriage Act, might also count as examples oflegislative entrenchment. See TRIBE, supra note 67, at 125-26 n.1; Larry Alexander &Saikrishna Prakash, Mother May I? Imposing Mandatory Rules of Statutory Interpretation, 20

CONST. COMMENT. 97, 98-99, 1o8 (2004); Posner & Vermeule, supra note 24, at 1697-99;see also Lockhart v. United States, 546 U.S. 142, 147-49 (2005) (Scalia, J., concurring)

(expressing the view that a statutory requirement of express reference for override should be

treated as nonbinding). For a conflicting view, see Nicholas Quinn Rosenkranz, Federal

Rules of Statutory Interpretation, 11s HARv. L. REV. 2085, 2117-20 (2002), which finds noentrenchment problem with congressional enactment of prospective rules of statutoryinterpretation.

79. For a useful description, see Serkin, supra note 3, at 898-99.

so. See Michael W. McConnell, Why Hold Elections? Using Consent Decrees To Insulate Policiesfrom Political Change, 1987 U. CHI. LEGAL F. 295.

81. See, e.g., Dana & Koniak, supra note 67, at 478-79 (discussing the permissibility ofgovernment contracts with private industry to raise revenue by "selling law-making

421


Supreme Court has been of two minds about the enforceability of suchcontracts. In some cases, the Court has taken the view that contracts betweengovernments and private parties should be fully enforceable, emphasizing thepublic-regarding benefits of contractual commitment, the risk of politicalopportunism, and the reliance interests of private actors.2 Even whileenforcing contracts, however, the Court has struggled to distinguish the anti-entrenchment principle that "one legislature is competent to repeal any actwhich a former legislature was competent to pass; and that one legislaturecannot abridge the powers of a succeeding legislature."8 This principle attainsprimacy in cases where the Court has refused to enforce government contracts,taking the view that the government cannot contract away sovereign authority.In those cases, the Court emphasizes the possibility of corrupt or imprudentcontractual obligations, the need for responding to changed circumstances,and, above all, the democratic imperative of contemporaneous self-governance.4

Scholars have embraced and amplified these normative concerns as appliedto legislative entrenchment more broadly. Critics of entrenchment argue thatcurrent legislatures will possess more information than past ones and thatdisallowing them from adapting to changed circumstances would lock inerroneous and anachronistic decisions.8 ' They also argue that legislative

authority"); Daniel R. Fischel & Alan 0. Sykes, Governmental Liability for Breach of Contract,i AM. L. & EcoN. REV. 313, 313-17, 326-32 (1999) (discussing differences between privateparties and the government with regard to available contract remedies); Gillian Hadfield, OfSovereignty and Contract: Damages for Breach of Contract by Government, 8 S. CAL. INTERDISC.L.J. 467, 467 (1999) (describing tension "between the power of government to bind itselfand future governments in contract and the freedom of a democratically elected legislatureto override the acts of a prior legislature in response to evolutions in judgment, information,or politics").

82. See, e.g., Trs. of Dartmouth Coll. v. Woodward, 17 U.S. (4 Wheat.) 518 (1819); New Jerseyv. Wilson, ii U.S. (7 Cranch) 164 (1812); Fletcher v. Peck, io U.S. (6 Cranch) 87 (1810).

83. Fletcher, io U.S. (6 Cranch) at 135; see also United States v. Winstar Corp., 518 U.S. 839, 873& n.19 (1996) (plurality opinion) (recognizing the principle "that 'a general law ... may berepealed, amended or disregarded by the legislature which enacted it,' and 'is not bindingupon any subsequent legislature,"' even while enforcing a government contract (quotingManigault v. Springs, 199 U.S. 473, 487 (1905))).

84. See, e.g., United States Tr. Co. v. New Jersey, 431 U.S. 1, 33 (1977) (Brennan, J., dissenting);Stone v. Mississippi, iol U.S. 814, 817 (1880) (holding that a state legislature cannot bargainaway its police power); W. River Bridge Co. v. Dix, 47 U.S. (6 How.) 507, 507 (1848)(holding that a state cannot contract away its eminent domain power); see also Fischel &Sykes, supra note 81, at 319 (explaining the "vexing and recurring problem concern[ing] thegovernment's ability to enter into long-term contracts" as conflicting with the constitutionalanti-entrenchment principle).

8s. See Eule, supra note 67, at 387; Roberts & Chemerinsky, supra note 77, at 1811-12.

422

125:4oo0 2015


entrenchment would exacerbate the damage that a badly motivated majority orone with extreme or aberrational preferences could inflict on the country."More generally, they argue that, like electoral entrenchment, legislativeentrenchment illegitimately undermines democratic accountability andmajority rule by disempowering legislatures from acting on the preferences ofcurrent majorities. As Julian Eule put it in his foundational argument againstlegislative entrenchment, "The fundamental . . . assumption of Americanpolitical life -that legislative action reflects current majoritarian preferences -could be finally laid to rest if shifting majorities were unable to alter priormajoritarian choices.",8

3. Common Denominators

Courts and scholars have treated electoral and legislative entrenchment astwo separate and distinct phenomena.8" The conspicuous difference betweenentrenching parties and politicians in office, on the one hand, and entrenchingenacted statutes and the policy decisions they embed, on the other, has struckmost observers as sufficient reason to place electoral and legislativeentrenchment in separate categories.

Upon closer inspection, however, the basis for this categorical distinctionbegins to blur. After all, elections matter in large part because they decide whocontrols the government and, consequently, the kinds of laws and policieslikely to be generated. Correspondingly, at least one important reason electoralentrenchment strikes many as problematic is that it permits politicians whohave lost popular political support to enact laws and policies that the medianvoter disprefers. If, for example, the Democratic Party can manipulate electionlaw to retain a majority of seats in a state legislature even after losingmajoritarian support, it might use its power to raise taxes on the rich or legalizemarijuana, despite the fact that most voters and citizens might prefer lowertaxes or oppose legalization. But legislative entrenchment can accomplish thevery same ends, and it raises the very same concerns. Suppose, in our example,that electoral entrenchment is impossible. In its fleeting moment ofmajoritarian ascendance, the Democratic Party, anticipating defeat at the pollsin the next election, might enact unrepealable laws raising taxes and legalizing

86. See Eule, supra note 67, at 388; Roberts & Chemerinsky, supra note 77, at 1809-11, 1813.

8s. Eule, supra note 67, at 405; see also Dana & Koniak, supra note 67, at 526-36 ("If majorityrule means anything, it means rule by the current majority and not by a majority of thepast.").

88. But see Klarman, supra note 4, at 504-07 & nn.63-69 (making the connection betweenlegislative and electoral entrenchment).

423


marijuana. Even after Republicans commanded a legislative and electoralmajority, they would be powerless to reverse these policy decisions. Thus, tothe extent that countermajoritarian policymaking is the driving concern,electoral and legislative entrenchment seem functionally and normativelyequivalent.8 ,

The same point holds in the reverse: when legislative entrenchment or itsequivalent is impossible, electoral entrenchment can serve as a substitute.Critics of legislative entrenchment worry that a narrow legislative majoritymight, for instance, enact a ban on capital punishment and, "knowing that thevoters will be angry and will want to elect pro-death penalty replacements,"entrench the statute against repeal.90 But these commentators might equallywell worry that the same precarious majority would use the tools of electionlaw- gerrymandering districts, disfranchising or defunding opponents, andthe like -to defeat their pro-death-penalty opponents and retain office. Eitherform of entrenchment would prevent capital punishment from being restoredby an opposed majority.

The simple point is that electoral entrenchment and legislativeentrenchment are substitutable strategies for accomplishing the same basicresult: locking in substantive policy outcomes. Entrenched policy outcomes canbe generated either indirectly through electoral entrenchment or directlythrough legislative entrenchment. From the perspective of both political actorsscheming to protect policy outcomes and citizens concerned with what thesepolicy outcomes will turn out to be, the bottom line is largely the same.9 '

ag. Commentators have recognized that government contracts can be used to lock in bothspecific policies and broader party platforms or coalitional agendas:

[A] political party in power might undertake to ensure the survival of its policiesagainst the contingency of future political defeat by entering contracts that couldmake changes in policy extremely expensive. Proregulatory forces might enterlong-term contracts with private entities for expensive regulatory services;antiregulatory forces might contract with the private sector to reimburse it for thecosts of any future changes in regulation; small-government proponents mightenter contracts promising compensation for any increase in taxes; big-government proponents might enter long-term employment contracts withgovernment workers. By making it difficult or impossible to change policies onceput in place, incumbent officials could thwart the possibility of democraticchanges to public policy.

Fischel & Sykes, supra note 81, at 338.

go. Roberts & Chemerinsky, supra note 77, at 1798.

91. A corollary observation is that theories of electoral entrenchment that emphasize theimportance of preserving robust partisan competition and preventing partisan lock ups donot really get to the heart of what is problematic about entrenchment. See Issacharoff &Pildes, supra note 26. Imagine that a temporarily ascendant Republican majority in Ohio

424

12S:400 201S


This is not to deny that voters care about the identity or party label of therepresentatives they elect to office for reasons independent of the policyoutcomes those representatives are likely to produce. Voters may want toreplace incumbent officeholders who are incompetent, corrupt, or personallydislikeable, regardless of the consequences for policymaking. To the extent thatvoters have policy-independent electoral preferences along these lines, thensome forms of electoral entrenchment will be viewed as problematic ongrounds that have no equivalent when it comes to legislative entrenchment.Consider, for example, a bipartisan gerrymandering scheme that generatessafe, noncompetitive districts for all incumbents but results in perfectlyproportional representation for the two major parties. Governance outcomesunder such a scheme would continue to reflect the policy preferences of theelectorate, and indeed more voters would find themselves in districtsrepresented by someone who shared their policy preferences and partyaffiliation than if districting were random. Yet incumbent officeholders whomight otherwise have been unseated would be safely entrenched in office.Many people might prefer a "fair" election, even if the only change would be toreplace those incumbents with co-partisans who would generate similar policyoutcomes.92

enacts a law imposing a wealth requirement for voting, effectively disfranchising all poorpeople in the state. The immediate electoral impact of such a law might be to depriveDemocratic candidates of any chance of winning state and federal offices that wouldotherwise have been competitive - the Republican Party would be entrenched in power. Theimmediate policy impact would be a correspondingly sharp turn to the right as the interestsof poorer voters were discounted, perhaps resulting in reduced spending on welfareprograms, inner-city public schools, and the like.

But the Republican ascendancy would not last forever. Democrats would quickly realizethat their only hope of securing or retaining office would be appealing to the median voterof this new, wealthier electorate. The Democratic Party in Ohio and individual Democraticofficeholders and candidates would presumably shift their platforms accordingly, movingthem close to the platforms of prevailing Republicans. Eventually, competition and somesort of rough parity between the parties would be restored, and the entrenchment ofRepublicans would come to an end. But- and here is the important point- this would becold comfort to supporters of the old Democratic platform and the predisfranchisementmedian voter. Whatever the party label of prevailing politicians in the new regime,governance outcomes would reflect the preferences of the new, wealthy electorate -more

closely resembling the initial Republican coalition, resulting in policies more closelyresembling the initial Republican platform. So long as that group of citizens and that set ofpolicies remained entrenched, the end of partisan entrenchment seems relativelyunimportant. Cf Stephanopoulos, supra note 37, at 299-300 (observing that the existence ofcompetition is no guarantee that electoral outcomes will align with voters' preferences).

92. Compare Samuel Issacharoff, Gerrymandering and Political Cartels, 116 HARv. L. REv. 593,612-30 (2002) (arguing that bipartisan gerrymanders represent a fundamental failure ofcompetitive democracy), with Nathaniel Persily, In Defense of Foxes Guarding Henhouses: The

425


Still, at least a large part of the reason citizens will object when a party,officeholder, or electoral faction retains power by means of electoralentrenchment is that the resulting governance outcomes are likely to deviatefrom their preferences. And at least as a first approximation, these outcomescan be generated equally well by entrenching them directly or by entrenchingthem indirectly, such as by entrenching their proponents in office. This linkagebetween electoral and legislative entrenchment reflects the simple fact thatpolitical power is primarily valuable because of what it can be used toaccomplish.

The linkage operates at a normative level as well. As we have seen, theperceived pathologies of entrenchment in both categories are nearly identical.Locking in parties, politicians, and policies alike threatens the democratic valueof rule by present majorities - replacing democratic responsiveness to popularpreferences with dead-hand control of the past and anachronistic ormaladapted governance outcomes. This should come as no surprise. If electoraland legislative entrenchment create similar functional outcomes, then to theextent these outcomes are viewed as democratically pathological, the diagnosiswill be the same for both.

II. FROM FORMAL TO FUNCTIONAL ENTRENCHMENT

As the previous Part surveyed, concerns about entrenchment in public lawhave been focused on formal legal rules that create impediments to politicalchange - rules of election law that increase the difficulty of replacingincumbents or prevailing political parties in office or rules about legislationthat increase the difficulty of revising enacted policies. But politicalentrenchment can also be accomplished without any shift in the legal rulesdirectly governing permissible processes of political change. As this Partdescribes, what electoral and legislative entrenchment accomplish formally andlegally can also be accomplished functionally and politically.

A. The Idea ofFunctional Entrenchment

Constitutional theory provides a useful point of entry to the distinctionbetween formal and functional entrenchment. Constitutional theorists havelong understood that formal, legal entrenchment is "neither necessary nor

Case for Judicial Acquiescence to Incumbent-Protecting Gerrymanders, n6 HARv. L. REv. 649,654-73 (2002) (emphasizing that noncompetitive electoral districts can both satisfy voterpreferences and generate proportional representation at the legislative level).

426

125:4oo0 2015


sufficient to create functional political entrenchment."9 3 Formal entrenchmentof the text does not prevent constitutional change, because constitutionalchange need not be channeled through the Article V amendment process.Indeed, as a practical matter, the most important mechanism of constitutionalchange has been through shifting interpretations or constructions of themeaning of the (formally unchanged) constitutional text by courts, politicalactors, and the public.94 Formal entrenchment in the constitutional text istherefore no guarantee of functional stability.

Conversely, constitutional theorists have emphasized that functionalstability need not depend on formal constitutional status. Thus, theorists havepointed to a number of norms outside the constitutional text that are treated inpractice as impervious to ordinary political contestation or change-and thusthat might be understood as functionally, even if not formally, constitutional.95

In Bruce Ackerman's view, for example, constitutional norms may be created orrewritten when the American public is roused to transcend ordinary politicsand engage in a higher-order form of deliberation about the public good. 6

These norms may float free of any particular legal text, or they may be codifiedin formally nonconstitutional statutes like the 1964 Civil Rights Act and the1965 Voting Rights Act.97 For Ackerman, what makes these normsconstitutional is not just their special democratic pedigree but also theirinvulnerability to ordinary political revision. Thus, Ackerman emphasizes that"an all-out assault on the Civil Rights Act, or the Voting Rights Act, could notoccur without a massive effort comparable to the political exertions that createdthese landmarks in the first place." 8 Similarly seeking to define constitutionallaw functionally rather than formally, Ernest Young concludes that the onlyinteresting and distinctive sense in which some legal norms should beconsidered constitutional is that they are "entrenched" against change.99

Among other examples, Young points to the Social Security Act's promise of

93. Levinson, supra note 17, at 697-98.

94. See generally JACK M. BALKIN, LIVING ORIGINALISM (011); STRAUSS, supra note 29.

95. Karl Llewellyn, writing in the 1930s, defined our "working [c]onstitution" as the set ofnorms and institutional arrangements that political actors treat as "not subject to abrogationor material alteration." Karl N. Llewellyn, The Constitution as an Institution, 34 COLUM. L.REV. 1, 28-29 (1934).

96. See 1 BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991); 2 BRUCE ACKERMAN, WE

THE PEOPLE: TRANSFORMATIONS (1998); 3 BRUCE ACKERMAN, WE THE PEOPLE: THE CIVIL

RIGHTS REVOLUTION (2014).

97. See Bruce Ackerman, The Living Constitution, 12o HARv. L. REV. 1737, 1757-93 (2007).

98. Id. at 1788.

99. Ernest A. Young, The Constitution Outside the Constitution, 117 YALE L.J. 408, 426 (2007).

427


government financial support in old age, which he plausibly predicts is lesslikely to be "fundamentally altered or abolished over the next ten years" thancanonical constitutional norms like the rights to burn an American flag or getan abortion."oo

Whatever protects Social Security, the Civil Rights Act, and perhaps other"superstatutes"'0 ' from revision or repeal, it is not any kind of formal barrier tochange. These statutes and the arguably "constitutional" norms they embodyare formally susceptible to repeal or revision through the ordinary Article I,Section 7 procedures. Nonetheless, as these theorists recognize, laws andpolicies can be protected by political barriers that may be every bit as difficultto overcome as constitutional barriers or other formal impediments tochange."o2 These laws and policies are functionally, even if not formally,entrenched.

But functional entrenchment is hardly limited to a small set of landmark,quasi-constitutional statutes. For example, critically assessing the prohibitionon legislative entrenchment, Posner and Vermeule observe that other types ofgovernment actions, beyond formal entrenchment, might share the purposeand effect of altering the downstream political environment in such a way as toincrease the costs of changing course, even to the point of practicalimpossibility. They offer the example of a legislature intent on entrenching apolicy against riding bicycles in a park: if barred from enacting an unrepealablestatute, the legislature might instead replace the existing concrete paths in thepark with bicycle-unfriendly gravel, effectively raising the downstream

1o. Id. at 427.

io. Taking a similar approach to Ackerman's and Young's, William Eskridge and John Ferejohnidentify as "America's Working Constitution" a set of "superstatutes," including the 1964Civil Rights Act, the Voting Rights Act, and the Endangered Species Act of 1973, whichcreated "entrenched governance structures and normative commitments." WILLIAM N.ESKRIDGE, JR. & JOHN FEREJOHN, A REPUBLIC OF STATUTES: THE NEw AMERICAN

CONSTITUTION 7 (2010).

102. In fact, as Young suggests, formal barriers may be much less effective in preventing politicaland constitutional change than functional ones. More strongly: legal constraints may bemeaningless in the absence of underlying political support. Recall the fear expressed byJames Madison and other designers of the U.S. Constitution that constitutional rights andrules would create mere "parchment barriers." THE FEDERALIST NO. 48, at 276 (JamesMadison) (Clinton Rossiter ed., 1999). Madison recognized that legal prohibitions wouldonly be meaningful if they could be made politically self-enforcing, by way of politicaldecision-making processes and institutions that would selectively empower decision makerswith the right interests and incentives. Generalizing the point, constitutional and otherforms of legal entrenchment always depend on the political entrenchment of the relevantlegal rules or their substance. See generally Levinson, supra note 17; infra notes 282-286 andaccompanying text.

428

125:400 2015


financial and political costs of bringing back the cyclists.o3 Here, a functionalmechanism of entrenchment substitutes straightforwardly for a formal one.o4

What theorists of subconstitutional entrenchment seem not to appreciate,however, is that examples like this are the rule, not the exception. Statutes andpolicies, as well as politicians and parties, can be entrenched just as readily byfunctional, political mechanisms as by formal, legal ones.

Indeed, political scientists, economists, and sociologists have generated avast and varied literature exploring the many different means by whichpolitical actors seek to insulate power holders and policies against downstreampolitical change without recourse to formal entrenchment devices. Thesefunctional entrenchment strategies take a number of different forms. Forillustrative purposes, we focus on three of the most general.'o One is the

103. Posner & Vermeule, supra note 24, at 1687.

104. In fact, upon closer inspection, a number of examples of legislative entrenchment featured inthe scholarly literature seem to involve functional without formal entrenchment. Theoristswho characterize the Gramm-Rudman Act and other framework statutes as impermissiblelegislative entrenchment recognize that there is no legal impediment that prevents asubsequent congressional majority from revising or repealing the original statute. To theextent framework statutes entrench policies or procedures against change, they do so byraising the political salience - and therefore, in some cases, the political cost - of reversal.Gramm-Rudman, for example, was designed to impose "substantial political risks" onsubsequent congressional majorities that chose not to comply by forcing them to overcomethe burden of "legislative inertia" and "to act in the full glare of the balanced budgetdebate." Kahn, supra note 67, at 2o5; see also Elizabeth Garrett, The Purposes of FrameworkLegislation, 14 J. CONTEMP. LEGAL ISSUES, 717, 748-53 (2005). The entrenchment of theSenate filibuster may be similar, if the Senate rules are interpreted to permit a currentmajority to lower the voting threshold for cloture. For that matter, if the remedy for breachof government contracts is limited to damages -as the Court has strongly implied it mustbe-then the only barrier to legal change is whatever political consequences flow frompaying that amount of money. See United States v. Winstar Corp., 518 U.S. 839, 885-87(1996) (plurality opinion). At the extreme, one could argue that the costs of entrenchmentcan always be boiled down to political costs, inasmuch as even a clear legal prohibition or ajudicial command could be ignored or overridden by political actors willing to suffer thepolitical consequences.

105. This is not meant to be an exhaustive list. Other methods and mechanisms of functionalpolitical entrenchment might be identified, and the various methods might be taxonomizedin various ways. For example, Paul Pierson develops a set of mechanisms of politicalstabilization and path dependence by analogy with the economic phenomenon of increasingreturns. PAUL PIERSON, POLITICS IN TIME: HISTORY, INSTITUTIONS, AND SOCIAL ANALYSIS 17-

53 (2004). Patterns of technology adoption, industrial location, and international trade havebeen explained as emerging from a path-dependent process through which slight initialadvantages snowball into irreversible market dominance. Increasing returns are commonlycreated by several features of the economic context: (1) large set-up or fixed costs, whichlead to lower marginal costs of producing additional units and create an incentive to stickwith the initial design; (2) learning effects, which increase the value of the product over

429


enactment of substantive policies that have the effect of strengthening politicalallies or weakening political opponents. A second, related strategy is to enactpolicies or programs that have the effect of changing the composition of thepolitical community-selecting in allies or selecting out opponents. A thirdstrategy is to shift the locus of political decision making, empowering adifferent governmental institution and consequently a different set of politicalactors and groups.

In the Sections that follow, we illustrate these common mechanisms offunctional political entrenchment with a number of examples, some drawnfrom existing work in the social sciences and others that we develop on ourown. Each of these examples involves a political strategy designed toaccomplish functionally what equivalent electoral or legislative entrenchmentstrategies might have accomplished formally.

B. Money and Mobilization

The most straightforward strategy of political entrenchment is toselectively empower one's allies or to selectively disempower one's enemies.One way of accomplishing this, of course, is by manipulating the legalframeworks governing elections and legislation. But another, and perhapsmore pervasive, way of achieving the same results is to engineer policyinitiatives that organize, mobilize, and enrich interest groups and otherconstituencies with a stake in defending one's preferred policies and theofficials who enacted them, or that demobilize or drain the resources of interestgroups and constituencies on the other side. In E.E. Schattschneider's well-known summation, "New policies create a new politics.",1o6 And new politics

time; and (3) coordination effects, including network externalities, which increase the valueof a product as more people use it and expect others to use it in the future. See generally W.BRIAN ARTHUR, INCREASING RETURNS AND PATH DEPENDENCE IN THE ECONOMY (1994). Theclassic example is the QVERTY typewriter keyboard. See generally Paul A. David, Clio andthe Economics of QWERTY, 75 AM. ECON. REV. 332 (1985). Pierson shows how each of thesefeatures has political analogues. The design of political institutions and policy regimes oftenentails high initial set-up costs. These institutions and regimes often inspire specific,nontransferable investments by various political actors and increase the power of theseactors to block change. The benefits of coordinating around an existing set of institutionsand policies lends further stability to these arrangements by creating equilibria in which nogroup can do better by withdrawing or contesting the status quo. See PIERSON, supra, at 22-

53, 142-57; see also Levinson, supra note 17, at 681-91 (elaborating on these and othermechanisms of political stability and entrenchment).

106. E.E. SCHATTSCHNEIDER, POLITICS, PRESSURES AND THE TARIFF 288 (1935); see also THEDASKOCPOL, PROTECTING SOLDIERS AND MOTHERS: THE POLITICAL ORIGINS OF SOCIAL POLICY

430

125:4oo0 2015


can be engineered in ways that entrench enacted policies or the parties orpoliticians that created them, through political feedback effects.1 o

7

The classic example to which we have been referring throughout is SocialSecurity, which created a vested interest group that was induced to rely on thebenefits of the program and strongly motivated to resist retrenchment. Prior tothe enactment of the Social Security Act, senior citizens in the United Stateswere neither a politically active nor particularly powerful constituency. ButSocial Security galvanized seniors, providing them with a focused motivationfor defending their benefits and the material and organizational resources totransform themselves into a formidable interest group. As a result, SocialSecurity became increasingly untouchable-the notorious "third rail" ofAmerican politics.,os

Other examples of political feedback effects focus on entrenchment by wayof disempowering enemies. For example, scholars have documented howairline deregulation reduced the economic and political cohesion of theindustry and therefore the political pressure that could be applied in favor ofrecartelization.o' Similarly, international trade agreements and free-tradepolicies tend to channel wealth and (consequently) political power away fromimport-competing interests and toward export interests and therefore to erodetheir own opposition while building their own support.'

Political actors may not always intend or anticipate these self-entrenchingpolitical feedback effects. But it would be remarkable if political actors were notattuned to the self-serving possibilities, and in at least some cases there is clearevidence that entrenchment was not just an unintended byproduct but part ofthe self-conscious design of particular programs and policies.

IN THE UNITED STATES 57-60 (1992) (describing how "[p]olicies [t]ransform [p]olitics").Schattschneider was generalizing from the example of the tariff:

By means of the protective system governments stimulate the growth ofindustries dependent on this legislation for their existence, and these industriesform the fighting legions behind the policy. The tariff likewise destroys interests.The losers adapt themselves to the new conditions imposed upon them, findthemselves without the means to continue the struggle, or become discouragedand go out of business. Is this not true, in varying degrees, of nearly all otherpolicies also? New polices create a new politics.

SCHATTSCHNEIDER, supra, at 288.

107. See PIERSON, supra note 105, at 30.

108. See ANDREA LOUISE CAMPBELL, How POLICIES MAKE CITIZENS: SENIOR POLITICAL ACTIvisMAND THE AMERICAN WELFARE STATE (2003); Paul Pierson, The New Politics of the WelfareState, 48 WORLD POL. 143, 144-47 (1996).

iog. See, e.g., PATASHNIK, supra note 12, at 179-80.

no. See ERIC A. POSNER, THE PERILS OF GLOBAL LEGALISM 53-54 (2009).

431


1. From Poll Taxes and White Primaries to Labor Law

The struggle for political power in the American South during the early andmiddle decades of the twentieth century is oft-invoked as a seminal example ofelectoral entrenchment. The Democratic Party during this period had amonopoly on Southern politics, and the Party itself was ruled by a faction ofwhite conservatives. This ruling faction was able to maintain its grip on theParty - and thus on Southern politics - only by disfranchising blacks and poorwhites."' To this end, Party leaders engineered the enactment of poll taxes andprohibitions on black participation in primary elections."2

But poll taxes and the white primary were not the only tools of electoralentrenchment. Labor law was another device used by conservative Democratsto fend off threats to their dominance. Those threats came from unions.Beginning in the late 1930s, the Congress of Industrial Organizations (CIO) -the labor movement's progressive wing -sent teams of organizers to the Southin an attempt to realign the Democratic Party."' Union efforts began withpolitical education and so-called "citizenship classes" that covered theprocedures for voting, the importance of voting, and tools for organizingcommunities to vote."4 The unions' political work also included voter

i%. Because blacks and poor whites almost always outnumbered white elites in any electiondistrict, the Democrats' strategy was a tenuous one: if even a small proportion of blacks orpoor whites could be registered and enabled to vote, power within an individual district, oracross the Party as a whole, could be shifted. As Patricia Sullivan concludes, "The inordinatepower enjoyed by southern members of Congress was dependent on a small electoraterestricted by race and class." PAT1ICIA SULLIVAN, DAYS OF HOPE: RACE AND DEMOCRACY INTHE NEw DEAL ERA 105 (1996).

m. See Issacharoff & Pildes, supra note 26, at 652-68. After Smith v. Allwright, 321 U.S. 649(1944), and the constitutional invalidation of the white primary, the poll tax became aprimary tool of black disfranchisement. As Steven Lawson writes, "Along with literacy testsand registration requirements, the tax ... dramatically sliced voter turnout and discouragedthe organization of political-party opposition." STEVEN F. LAwsON, BLACK BALLOTS: VOTINGRIGHTS IN THE SOUTH, 1944-1969, at 55 (1976).

113. Between 1937 and 1939, for example, the CIO's Textile Workers Organizing Committeespent two million dollars and had six hundred organizers in the field working on thisSouthern strategy to "recast the Democratic Party." STEVEN FRASER, LABOR WuL RULE:SIDNEY HILLMAN AND THE RISE OF AMERICAN LABOR 387 (1991). When, in 1943, the CIO

established its Political Action Committee (the CIO-PAC), it too focused on Southernpolitical organizing. CIO-PAC established regional offices in Virginia, North Carolina,South Carolina, Georgia, Alabama, Tennessee, Mississippi, Florida, Louisiana, Texas,Missouri, Kansas, Oklahoma, and Arkansas. JOSEPH GAER, THE FIRST ROUND: THE STORYOF THE CIO POLITICAL ACTION COMMITTEE 230 (1944).

114. See, e.g., Robert Korstad & Nelson Lichtenstein, Opportunities Found and Lost: Labor,Radicals, and the Early Civil Rights Movement, 75 J. AM. HIST. 786, 792-93 (1988).

432

125:40 0 2015


registration drives, conducted by both black and white fieldworkers and aimedat both poor white and eligible black voters. 5 Perhaps most importantly,unions developed campaigns to pay poll taxes on behalf of voters who couldnot afford them." 6

The goal of these efforts was straightforward: to organize black andworking-class white voters and get them to the polls to "rid the DemocraticParty of conservatives."n7 The union efforts produced significant results. Afterone successful poll tax campaign among oil workers in a Texas congressionaldistrict-a campaign that led to registration rates "higher than ever knownbefore in the region" -the incumbent Democrat withdrew from his reelectioncampaign."' In Huntsville, Alabama, where the CIO paid poll taxes for fourthousand workers in a population of approximately 13,500, the fiercelyconservative incumbent Joe Starnes was defeated."9

Conservative incumbents understood the political threat the union posed,and "[t]hroughout the World War II era, Southern congressmen, newspapers,and business leaders [railed against CIO] efforts to mobilize black voters andpoor whites."2 o Incumbents also understood that, just as poll taxes couldneutralize political opposition from blacks and poor whites, so too couldrestrictive labor legislation. Thus, nearly as soon as CIO organizers arrived inthe South, Southern state legislatures became "hotbeds of antilabor

115. See SULUIVAN, supra note iii, at 173.

116. In 1943, for example, the CIO led a poll tax drive in Martin Dies's Texas congressional

district that resulted in a twenty-five to thirty percent increase in voter registration rates. In

1944, the director of the CIO-PAC's Southern region reported that seventy-five thousand

previously unregistered workers had registered and paid poll taxes in eight Southern states.

ROBERT A. GARSON, THE DEMOCRATIC PARTY AND THE POLITICS OF SECTIONALISM, 1941-

1948, at 76 (1974).

117. GILBERT J. GALL, THE POLITICS OF RIGHT To WORK: THE LABOR FEDERATIONS AS SPECIAL

INTERESTS, 1943-1979, at 30 (1988).

118. MATTHEW JOSEPHSON, SIDNEY HILLMAN: STATESMAN OF AMERICAN LABOR 610-11 (1952).

ii9. GARSON, supra note 116, at 76; see also William H. Riker, The CIO in Politics 1936-1946, at

310-11 (1948) (unpublished Ph.D. dissertation, Harvard University) (on file with authors).During those same 1944 primaries, incumbents in South Carolina, Missouri, Arkansas,Alabama, and Texas "all lost to PAC-endorsed candidates." FRASER, supra note 113, at 514.

12o. ROBERT H. ZIEGER, THE CIO: 1935-1955, at 230-31 (1995). "The CIO-PAC, complainedNorth Carolina senator Josiah Bailey in 1944, 'will seek to purge us and every other self

respecting and honest man who runs for office."' Id. at 230. The CIO's organizing efforts

"scared the daylights" out of South Carolina Senator Ellison Durant "Cotton Ed" Smith,BRYANT SIMON, A FABRIC OF DEFEAT: THE POLITICS OF SOUTH CAROLINA MILLHANDS, 1910-

1948, at 197 (1998), and, during the 1948 Texas Democratic primary, a segregationist

congressman received a report that "'Negroes outnumbering whites almost 3-1 were led by

CIO and AF of L leaders,' and this mutual cooperation forced the retreat of the Dixiecrats,"

LAWSON, supra note 112, at 127.

433


legislation."... After the CIO fiercely but unsuccessfully opposed the reelectionof Texas Governor Wilbert Lee O'Daniel, O'Daniel pushed for legislation thatoutlawed much core union activity.'" Two years later, Texas passed theManford Act which prohibited unions from charging dues that would "create afund in excess of the reasonable requirements of such union" - in other words,a fund that could be used for political activity.' For good measure, the TexasAct also flatly forbade unions from making political contributions7" Theselaws were replicated in states across the South.2 s

The Democratic push for restrictive labor laws was not confined to statelegislatures; Southern Democrats who were CIO targets also took their effortsto Congress."' Thus, in 1943, Howard Smith of Virginia co-sponsored the WarLabor Disputes Act, a law that banned political contributions by unions.'Several Southern Democrats who had been targeted by CIO organizing effortsalso used their positions on the House Un-American Activities Committee toinitiate investigations into the CIO's political activities.

Of course, restrictive labor laws served purposes beyond politicalentrenchment. Union activity in the South threatened entrenched economicinterests as directly as it threatened entrenched political leaders. But the dualpurpose of the Southern Democratic approach to labor policy does notdiminish its self-interested political dynamic. As George Norris Greendescribed Texas Democratic politics during this era, the Party establishment"not only feared the economic disadvantages of unionism for Dixie'scorporations, but also opposed Northern labor's encroachments in the highcouncils of the Democratic Party.""'

The strategic use of labor law as a political entrenchment mechanismbecame well established in American politics, with lasting effects on both laborlaw and political power. Perhaps the broadest and most powerful legislative

121. SULLIvAN, supra note iii, at 188.

122. Murray Emanual Polakoff, The Development of the Texas State CIO Council 47 (1955)(unpublished manuscript) (on file with authors).

123. 1943 Tex. Gen. Laws 180.

124. See Labor Laws and Court Decisions, 56 MONTHLY LAB. REv. 941, 942 (1943).

125. HARRY A. MILLIS & EMILY CLARK BROWN, FROM THE WAGNER ACT To TAFT-HARTLEY 323(1950).

126. See TRACY RooF, AMERICAN LABOR, CONGRESS, AND THE WELFARE STATE 1935-2010, at 28-

30 (2011); SuLLIvAN, supra note 111, at 174.

127. MIILIS & BROWN, supra note 125, at 346.

128. See JOSEPHSON, supra note 118, at 6o8-io.

129. GEORGE NORRIS GREEN, THE ESTABLISHMENT IN TEXAS POLITICS: THE PRIMITIVE YEARS,

1938-1957, at 61 (1979).

434

125:40 0 2015


attack on union political activity came in 1947 when a coalition of SouthernDemocrats and Northern Republicans, led by Senator Robert Taft of Ohio,passed the Taft-Hartley Act. That statute imposed a wide range of newrestrictions on unions, including major restrictions on union political activityand a complete federal ban on closed-shop agreements- thereby invalidatingwhat had been the unions' primary funding mechanism.3 o SouthernDemocrats in Congress were nearly unanimous in their support for Taft-Hartley.'' O'Daniel, the Texas Governor who had fought the CIO in statepolitics, was now a U.S. Senator and was particularly open about the politicalimportance of the Act. As he stated on the Senate floor:

When Senators are talking about the closed shop, they are talking aboutthe very heart and soul of the control of our American form ofgovernment, because it is the closed shop which siphons off from thetaxpayers and the honest laboring people of the country, hundreds ofmillions of dollars. This is done for the specific purpose of defeating thereelection of any Member of Congress who opposes the labor-leaderracketeers, and for the political purposes of using this money that isgained by virtue of the closed shop to elect to the Senate and to theHouse of Representatives men who will do the bidding of the laborleader racketeers.... The situation is a political one."'

130. See generally Sean Farhang & Ira Katznelson, The Southern Imposition: Congress and Labor inthe New Deal and Fair Deal, 19 STUD. AM. POL. DEV. 1, 16-17 (2005).

131. The political entrenchment function of the Taft-Hartley Act was also an importantmotivation for Senator Taft. Going into his 1944 reelection campaign, Taft believed that hisposition was secure, but the campaign became bitterly contested - and closely fought - whenthe CIO-PAC entered the race and gave its support to Taft's opponent. See JAMES T.PATTERSON, MR. REPUBLICAN: A BIOGRAPHY OF ROBERT A. TAFT 278 (1972). After theelection, Taft placed the blame for his near loss on the CIO, and in particular on its effort tomobilize "labor and the Negroes." Letter from Robert A. Taft to David S. Ingalls (Nov. 10,1944), in 2 THE PAPERS OF ROBERT A. TAFT, 1939-44, at 609, 609 (Clarence E. Wunderlin,Jr. ed., 2001). As Taft's biographer tells it, the 1944 campaign "left [Taft] grimly resolved tocurb the power of organized labor in the future." PATTERSON, supra, at 278. Returning to theSenate, Taft thus agreed to chair the Labor Committee -instead of the more prestigiousFinance Committee, which he had originally favored-in order to pass the antilaborlegislation that ultimately become the Taft-Hartley Act. Id. at 337-39.

132. 93 CONG. REC. 4,888-89 (1947) (statement of Sen. O'Daniel). Another example comes fromWisconsin in the 1950s. After labor unions mounted a substantial political challenge to theincumbent Republican governor-providing fifty-five percent of the Democraticchallenger's campaign contributions -Wisconsin Republicans took notice and introduced abill to "ban ... any political activity whatsoever by unions or their officers." Catlin Bill IsSigned; Kohler Praises Law, MILWAUKEE J., May 21, 1955, at i. As enacted, the Catlin Act"severely reduced labor's contribution to candidates of its choice and deprived the

435


Labor law continues to serve this electoral entrenchment function incontemporary politics. As we noted at the outset, because unions are criticalinstitutional supporters of the contemporary Democratic Party, underminingthe efficacy of labor unions is a well-understood means by which incumbentRepublican leaders can increase their reelection prospects."' Recall the recenteffort by Wisconsin Republicans to suppress opposition by amending thestate's labor laws to restrict the collective bargaining rights of public employeesand to prohibit public unions from collecting dues through payrolldeductions." Acting on the same motivation, Republican-dominated states

struggling Democrats of a major source of revenue." ROBERT W. OZANNE, THE LABOR

MOVEMENT IN WISCONSIN: A HISTORY 147 (2011). Democrats at the time recognized theintent and potential effect of the law, arguing that it was "intended to cripple theDemocratic Party and fasten one party rule on the state." Catlin Bill Is Signed, supra, at 7.The bill's sponsor did not disagree, "hail[ing] the passage of his act as a high water mark forthe Republican Party: [t] he power of labor unions to influence elections with money fromtheir treasuries was over." William R. Bechtel & Kenneth Fry, Catlin Act Vote Shows Big StatePolitics Shift, MILWAUKEEJ., May 4, 1959, at 5.

133. As J. David Greenstone famously observed, unions have been the "nationwide electoralorganization of the national Democratic Parry." J. DAVID GREENSTONE, LABOR IN AMERIcANPOLITICS Xiii (i969). While the relative strength of unions has declined in recent years, theirposition as a central force in Democratic politics remains stable. In the 2012 election cycle,for example, unions contributed one hundred forty-three million dollars to parties andcandidates; ninety-one percent went to Democrats. In the 2010 cycle, of the top five highest-spending nonparty organizations, the only organization that supported Democraticcandidates was a labor union. See 2010 Outside Spending, By Groups, CTR. FOR RESPONSIVE

PoLITICS, at xiii, http://www.opensecrets.org/outsidespending/summ.php [http://perma.cc/9VY3-N6LW]; see also Hendrik Hertzberg, Union Blues, NEW YORKER (Mar. 7, 2011),http://www.newyorker.com/talk/comment/2o1/o3/o7/11o3o7taco-talkhertzberg [http://perma.cc/9CJ7-9EGL]. And union efforts continue to impact election results: in the 20o8presidential elections, union membership increased by twelve percentage points thelikelihood that a voter would vote for Barack Obama, and unions boosted Obama's overallnational vote share by more than a full point. Nate Silver, The Effects of Union Membershipon Democratic Voting, N.Y. TIMES: FIVETHIRTYEIGHT (Feb. 26, 2011, 7:oo AM),http://fivethirtyeight.blogs.nytimes.com/2oll/o2/26/the-effects-of-union-membership-on-democratic-voting [http://perma.cc/7NGF-C3XC]. In state and local races, the union effectcan be even more pronounced.

134. The effect of these restrictions on public sector unions' ability to operate either as economicor political actors has been profound: union representation among Wisconsin's publicemployees dropped from 53.4% to 37.6% in the two years following the legislation'senactment. See Amanda Becker, U.S. Union Membership Steady at 11.3 Percent in 2013: LaborDepartment, REUTERS (Jan. 24, 2014), http://www.reuters.com/article/2014/ol/24/us-usa-labor-membership-idUSBREAoNMQ20140124 [http://perma.cc/H9MG-CJAD].

Following Wisconsin's lead, the Republican-controlled Ohio Legislature enacted asimilar law in 2011. See S. 5, 129th Gen. Assemb., Reg. Sess. (Ohio 2011). Indiana and

436

125:4oo0 2015


across the country have begun to enact so-called paycheck protection bills thatprohibit traditional methods of union dues collection.' Commenting on thesedevelopments, Steve Fraser and Joshua Freeman have observed: "[W]hat weare seeing is a partisan strategy to defund the Democratic Party, which hasreceived massive amounts of money from the union movement in recent years,especially from public sector unions.",, 6

The discussion so far highlights the ways that incumbents can usesubstantive policy to neutralize political opposition, in much the same way thatpoll taxes and primary rules were used to neutralize opposition. But policy canjust as easily be used to selectively mobilize political support. From an electoralentrenchment perspective, these are interchangeable tactics: both are ways ofusing the power of incumbency to shift the rules of the political game inincumbents' favor.

Thus, while in recent decades Republicans have sought to use their officesto undermine union strength in order to neutralize Democratic opposition,Democratic officeholders have just as aggressively sought to bolster unions inorder to shore up their own electoral prospects. At the federal level, forexample, Congress recently debated the Employee Free Choice Act (EFCA), abill that would have made private-sector unionization substantially easier."'Nearly the entire Democratic caucus in both the House and Senate supportedthe bill.,,8 While there were legitimate labor-policy reasons to support EFCA,Democrats could not have missed the possibility that increasing unionizationrates and the political power of organized labor would improve their electoralprospects. Critics certainly highlighted this feature of the proposed legislation.Writing in Labor Watch, W. James Antle put it this way:

[T]he Democrats and the labor unions have a symbioticrelationship . . . . The unions help the Democrats gain power, through

Michigan passed right-to-work statutes in 2012. See IND. CODE. § 22-6-6-8 (2012); 2012

Mich. Pub. Acts 348 (private sector); 2012 Mich. Pub. Acts 349 (public sector).

135. See, e.g., Gordon Lafer, The 'Paycheck Protection' Racket, EcON. POL'Y INST. (Apr. 24, 2013),http://s4 .epi.org/files/2013/paycheck-protection-racket-tilting-political.pdf [http://perma.cc/Q 325-XQY5].

136. Steve Fraser & Joshua B. Freeman, In the Rearview Mirror: A Brief Histoty of Opposition toPublic Sector Unionism, 20 NEw LAB. F. 93, 96 (2011).

137. See generally Benjamin I. Sachs, Enabling Employee Choice: A Structural Approach to the Rulesof Union Organizing, 123 HARY. L. REv. 655, 664-72 (2010).

138. Though not the entire caucus. For example, at least two Democratic senators from stateswith very low union density -Blanche Lincoln of Arkansas and Ben Nelson of Nebraska -did not support EFCA. See, e.g., Steven Greenhouse, Fierce Lobbying Greets Bill To HelpWorkers Unionize, N.Y. TIMES (Mar. lo, 2009), http://www.nytimes.com/2oo9/o3/11/business/iilabor.html [http://perma.cc/F6KR-L8G9].

437


their volunteers and their financial contributions. The Democratsreturn the favor by enhancing the unions' clout and trying to reversetheir membership's decline. This in turn means more dues with whichto help elect Democrats. The cycle continues . . . ."9

Likewise, at the state level, while Republicans in Ohio and Wisconsin havemoved to dismantle public sector unions, Democratic governors andlegislatures in states like Illinois, California, Oregon, and Iowa have moved toexpand union rights to new groups of public employees.140 These newlyunionized workers will undoubtedly provide a valuable source oforganizational and financial support for the Democratic governors andlegislators who enabled their organization. Commenting critically on one suchlaw, George Will put the point this way: "[T]he purpose of such systems is toenable unions to siphon away, in dues, a portion of [employees'] pay, some ofwhich becomes campaign contributions for the political party that created thesystem.""' As Will and others have recognized, labor law affects politicalpower and is a potentially powerful mechanism of political entrenchment.

2. From Campaign Finance Reform to Tort Reform

In election law scholarship and Supreme Court case law, campaign financereform is widely suspected of being another mechanism of incumbent andpartisan entrenchment. But formal campaign finance rules are not the only waythat incumbents and temporarily dominant parties can shore up their financialadvantage. Substantive policymaking can also be used to tilt the campaign-finance playing field. A clear contemporary example is tort reform. Triallawyers provide a significant portion of the funds relied on by Democratic

139. W. James Antle III, A Piece of the Action: Labor Expects Much from the Next Congress, NewAdministration, CAP. RES. CTR. 5 (Dec. 2008), http://capitalresearch.org/wp-content/uploads/2o13/o7/LW12o8.pdf [http://perma.cc/2FPM-ANHY].

140. For Ohio and Wisconsin, see Becker, supra note 134. For Illinois, California, Oregon, andIowa, see generally Benjamin I. Sachs, Labor Law Renewal, 1 HARv. L. & POL'Y REV. 375, 382-87 (2007).

141. George Will, Siphoning Compensation from Caregivers to Unions for Political Contributions,MISSOULIAN (Jan. 21, 2014), http://missoulian.com/news/opinion/columnists/syndicated/george-will-siphoning-compensation-from-caregivers-to-unions-for-political/article

4a57ca2a-82a2-nle3-b2dd-ool9bb2963f4.html [http://perma.cc/2TVC-HMUB]. Will's critique isnot precisely accurate. Union dues cannot be used for political "contributions" tocandidates, but they can be used to fund independent expenditures made on behalf ofcandidates, so long as the dues-paying union member does not object to such use. SeeBenjamin I. Sachs, Unions, Corporations, and Political Opt-Out Rights After Citizens United,112 COLUM. L. REv. 8oo, 809-19 (2012).

438

125:4oo0 2015


candidates in both state and federal elections."' Trial lawyers, in turn, rely onjury awards to generate the income they channel to Democratic candidates.'43

Consequently, legal reforms that reduce jury awards are an effectivemechanism for staunching the flow of funds to Democrats-and thus anattractive entrenchment device for Republicans.

This calculus contributed to the emergence of tort reform as a central plankin the Republican Party's platform starting in the 199os.'4 To be sure,Republicans supported tort reform for reasons other than partisan politicaladvantage; tort reform is a policy goal embraced on the merits by many interestgroups (and voters)."s But it is impossible to miss the fact that a significantpart of the attraction of tort reform for Republicans was the potential fordefunding their Democratic opponents.

Certainly Karl Rove did not miss it. In his early years as a Texas politicalconsultant, Rove presciently anticipated the political potential of tort reform.Beginning in the late 198os and continuing into the 1990s, Rove worked toelect Republican justices to the Texas Supreme Court with an eye towardreducing the size and frequency of jury awards. 6 When George W. Bushbecame Governor, Rove also pushed for significant legislative tort reform.47 Asone of Rove's longtime journalistic observers recounted, Rove decided to "runwith tort reform" in part because he thought the issue would play well with theTexas electorate, but also because he understood tort reform's partisanpotential:

142. See Sachs, supra note 141.

143. See, e.g., Timothy P. Carney, Trial Lawyer Industry Tries To Buy a DemocraticMajority, WASH. EXAMINER (Oct. 22, 2014), http://www.washingtonexaminer.com/trial-lawyer-industry-tries-to-buy-a-democratic-majority/article/2555105 [http://perma.cc/9RK5-3WKS]. This phenomenon is nothing new. See, e.g., Leslie Wayne, Trial LawyersPour Money into Democrats' Chests, N.Y. TIMES (Mar. 23, 2000), http://www.nytimes.com/2000/03/23/us/trial-lawyers-pour-money-into-democrats-chests.html [http://perma.cc/K4NB-8CAT].

144. See, e.g., Wayne, supra note 143.

145. See The Effects of Tort Reform: Evidence from the States, CONG. BUDGET OFF. 2-3 (June2004), https://www.cbo.gov/sites/default/files/lo8th-congress-2003-2004/reports/report_2.pdf [https://perma.cc/L3RB-PRNF] (presenting arguments for and against tort reform onthe merits).

146. See, e.g., S.C. Gwynne, Genius, TEX. MONTHLY (Mar. 2003), http://www.texasmonthly.com/story/genius [http://perma.cc/8SVB-DRAN]; Frontline, Karl Rove-the Architect:Interview: Sam Gwynne, PBS (Apr. 12, 2005), http://www.pbs.org/wgbh/pages/frontline/shows/architect/interviews/gwynne.html [http://perma.cc/GSL2-88W5] [hereinafterFrontline, Guynne].

147. Mimi Swartz, Left Behind, TEX. MONTHLY (Nov. 2011), http://www.texasmonthly.com/story/left-behind [http://perma.cc/3XGN-8RDD].

439


[I]t happened in the '8os that the major financing of the DemocraticParty in Texas ... began to be done by trial lawyers. If you looked atthe biggest givers to the Texas Democratic Party in the '8os and the'90s, you would see at the top of that list trial lawyers. So [tort reform]became this giant pitched battle, because it wasn't just necessarily aboutthe kind of verdicts and the ease with which someone might get averdict for a plaintiff, but it was also about the back end, which was thefinancing of the entire Democratic Party.

... It's a battle for the soul of Texas politics because it's a battle for themoney, the lifeline money of Democrats . . . .

Following Rove's successful use of the strategy in Texas, the RepublicanParty adopted tort reform as a national cause. In 1994, Newt Gingrich includedin his Contract with America a proposed bill called the Common Sense LegalReform Act,49 which would have preempted much of state tort law andimposed a federal punitive damages cap of two hundred and fifty thousanddollars in products liability cases.so The next year, the Republican-controlledHouse of Representatives enacted major restrictions on medical malpracticeawards.s1

Although elected leaders refrained from speaking openly about the politicalimplications of tort reform, Republican activists were less circumspect."s2 In1994, for example, Grover Norquist published a prominent essay arguing themerits of reform.'s Emphasizing trial lawyers' importance to the DemocraticParty, Norquist asserted that "[t]he political implications of de-funding thetrial lawyers would be staggering."'54 By 1999, with the presidential election of2000 looming, Norquist reiterated his case, arguing that even "[m]odest tort

148. Frontline, Gwynne, supra note 146.

149. Michael L. Rustad, Nationalizing Tort Law: The Republican Attack on Women, Blue CollarWorkers and Consumers, 48 RUTGERS L. REv. 673, 674 (1996).

iso. Id. at 675.

151. Id. at 68o.

is2. See generally STEPHANIE MENCIMER, BLOCKING THE COURTHOUSE DOOR: HOW THE

REPUBLICAN PARTY AND ITS CORPORATE ALLIES ARE TAKING AWAY YOUR RIGHT To SUE 97(2006) ("Republican activists began to talk openly about attacking lawyers because of theirpivotal role in funding Democratic politics.").

153. Grover G. Norquist, A Winning Drive, AM. SPECTATOR (Va.), Mar. 1994, at 60. The essaydiscusses the political benefits to Republicans of defunding the trial lawyers in the contextof a reform bill that would have enabled individuals to waive the right to sue for pain-and-suffering damages in order to secure reduced automobile insurance premiums.

154. Id. at 61.

440

125:4oo 2015


reform .. . would break the trial lawyers, second only to the unions as a sourceof funds for the left."1 5 5

Here again, the same policy that neutralizes political opposition can alsomobilize political support. Tort reform seems to have played this dual role forthe Republicans in the 1990s and 2000s: not only did it impede the ability oftrial lawyers to finance Democratic candidates, it also protected Republicanbusiness constituencies from high-dollar tort judgments and thus incentivized(and better enabled) this constituency to act as funders for the RepublicanParty. As John Podesta, White House Chief of Staff under Bill Clinton, told theWashington Post: "Why would you make [tort reform] the cause c6ldbre? ...It's important to them in both directions, both in organizing core elements oftheir business and doctor communities, and at least undermining a financialbase of the Democratic Party. ,,i6

C. Shaping the Political Community

Another well-documented entrenchment technique is for incumbentleaders to use the power of their offices to shape their own polities in such away as to ensure their lasting support. Election law scholars have been attentiveto this possibility in the context of districting, which presents politicians with

155. Grover G. Norquist, Winner Takes All: The 2ooo Elections Will Decide the Democrats' Future,AM. SPECTATOR (Va.), Apr. 1999, at 66, 67. In addition to trial lawyers, Norquist named"labor unions" and "Big City machines" as the other two pillars of the Democratic Party. Id.at 66. Following the election of President Bush, the Republican leadership did indeed moveto enact a series of tort reform measures. Again, neither Bush nor any administrationofficials spoke of the reforms as mechanisms of entrenchment. But others did: for example,Washington Post reporter Thomas B. Edsall wrote of the GOP's tort reform efforts that

"[t]he drive to limit court-awarded damages in civil lawsuits ... is usually framed as acontest between accident victims' rights and reasonable constraints on corporate behavior.Increasingly, however, the battle is deeply partisan, as conservative groups try tomobilize the political right and cripple a key Democratic constituency, trial lawyers."Thomas B. Edsall, Battle over Damage Awards Takes a More Partisan Turn, WASH.

POsT (Aug. 1o, 2003), http://www.washingtonpost.com/archive/politics/2oo3/o8/o/battle-over-damage-awards-takes-a-more-partisan-turn/ade8d300-94oe-4e4a-86ba-dao2c437e9e5[http://perma.cc/M96P-MEX8].

156. Thomas B. Edsall & John F. Harris, Bush Aims To Forge a GOP Legacy: Second-TermPlans Look To Undercut Democratic Pillars, WASH. POST (Jan. 30, 2005),http://www.washingtonpost.com/wp-dyVarticles/A47559-2oo5Jan29.html [http://perma.cc/W7CN-HW27]. Similarly, Ed Lazarus, a Democratic strategist, described tort reform asa "double header" because it worked both to "defund the Democratic Party" -by choking offtort damage awards-and to provide increased support to the Republicans by motivatingdonations by those industries negatively impacted by tort awards, including thepharmaceutical industry. Edsall, supra note iss.

441


the valuable opportunity to choose their voters. But here again, there arefunctional substitutes to gerrymandered districts. Substantive programs andpolicies can also be used to reshape politics in self-reinforcing ways byincreasing the number of proponents relative to opponents. For example,municipal gun control or antismoking ordinances will predictably gain politicalsupport over time as gun owners and smokers either give up their firearms andcigarettes or exit the jurisdiction, either way resulting in a higher percentage ofunarmed and nonsmoking supporters of the relevant policy and the officialswho promulgated it.' 7 Laws permitting more immigration or providing for thebetter treatment of immigrants will be similarly self-reinforcing, as a greaternumber of immigrants exercise more political power for the benefits of theirsuccessors.1ss

Although some selection effects along these lines will be unintentional,strategic politicians have every incentive to manipulate policy for the purposeof shaping their electorates and entrenching their hold on power. A famousexample is the "Curley Effect," described by Edward Glaeser and AndreiShleifer.'59 James Michael Curley, who served four terms as the mayor ofBoston during the first half of the twentieth century, was supported by apolitical base of lower-income Irish residents but opposed by Boston'swealthier Anglo-Saxon voters. To increase his own electoral prospects, Curleywas interested in keeping poor Irish in the city and in encouraging the wealthyAnglo-Saxons to leave. With no formal immigration law at his disposal, Curleyinstead used his control over public projects, patronage, zoning laws, and thelike to make things as hospitable as possible in Boston for his own voters andas uncomfortable as possible for his opponents."6

Selecting a supportive constituency ensured these leaders' political survivaland thus the continuation of their broader policy agendas.161 In the remainder

157. See Eugene Volokh, The Mechanisms of the Slippery Slope, 116 HARV. L. REV. 1026, 1116-18

(2003). Given Tiebout sorting dynamics, these examples can be generalized to differentkinds of policy decisions, especially at the municipal level where exit is less costly.

158. See id. at 118-19.

159. Edward L. Glaeser & Andrei Shleifer, The Curley Effect: The Economics of Shaping theElectorate, 21 J.L. EcoN. & ORG. 1 (2005).

16o. See id. at 10-12.

161. Less strategic politicians may be victimized by political selection effects running in theopposite direction. See, e.g., Shaila Dewan, Gentrification Changing Face of New Atlanta,N.Y. TIMES (Mar. 11, 2006), http://www.nytimes.con/2006/o3/11/national/iiatlanta.html[http://perma.cc/2ZZM-63CV] (quoting a community leader as saying that AfricanAmerican mayors have "cut [their] own throat[s]" by encouraging gentrification that hasdecreased the percentage of black voters in the city). Daron Acemoglu and James Robinsongeneralize this threat to many historical and political contexts to explain why power holders

442

125:4oo 2015


of this Section we consider two additional cases in which parties and politicianshave quite clearly and self-consciously pursued entrenchment through theselection of a favorable electorate. The first example is the admission of newstates in a federal system. Because each such admission threatens to shift thebalance of overall federal power, those decisions are made with an eye tomaintaining control by the currently dominant federal party. The secondexample is immigration policy, through which incumbents literally define theirown polities.

1. From Gerrymandering to State Admissions

In a federal system, the admission of new states can shift the balance ofnational power in the direction favored by the representatives of the newlyadmitted state. Consequently, current legislative majorities can cement theirhold on power by selectively admitting, or not admitting, new states. As wewill show, this dynamic, just like gerrymandering, can be bipartisan orpartisan. In the bipartisan form, legislatures lock in a policy status quo bydesigning state admissions policy in a way that maintains the current balanceof partisan power. In the partisan form, a dominant party advances its ownpolicy agenda and insulates that agenda against subsequent change bymanipulating the admissions policy.

The antebellum period provides a clear instance of bipartisan agreement onstate admissions for the purpose of maintaining the policy status quo. In theantebellum era, the relevant policy was slavery, and disputes over stateadmissions were primarily proxy fights in the sectional battle over slavery."'During much of this period, there was an equal balance of power betweenNorthern and Southern states in the Senate, which enabled either section toblock policies they opposed.6 3 But the balance of power meant that each state

often oppose economic development, out of fear that the resulting shifts in economic andpolitical power will lead to their downfall. See ACEMOGLU & ROBINSON, supra note i, at 83-91.

162. See, e.g., DAVID M. POTTER, THE IMPENDING CRISIS: 1848-1861, at 33 (1976); see also NolanMcCarty et al., Congress and the Territorial Expansion of the United States, in PARTY, PROCESS,AND POLITICAL CHANGE IN CONGRESS: NEW PERSPECTIVES ON THE HISTORY OF CONGRESS

392, 398 (David W. Brady & Mathew D. McCubbins eds., 2002) ("The admission of newstates during the antebellum period was tied to the conflict over slavery . . . ."). See generallyDON E. FEHRENBACHER, THE DRED SCOTT CASE: ITS SIGNIFICANCE IN AMERICAN LAW AND

POLITICS 152-87 (1978) (describing the events leading up to the Kansas-Nebraska Act of1954).

163. See Barry R. Weingast, Political Stability and Civil War: Institutions, Commitment, andAmerican Democracy, in ANALYTIC NARRATIVES 148, 151 (Robert H. Bates et al. eds., 1998).

443


admission could shift political control of the Senate, and thus control overnational policymaking, to one region or the other.

With the Senate evenly divided, neither North nor South could dominateadmissions politics, but each side could ensure the continuation of the statusquo. The result was a series of political compromises in which slave and freestates were admitted in pairs. Thus, at the time of the debate over theadmission of Missouri as a slave state in 1819, there were eleven northern statesand eleven southern states in the Union. 6 , Missouri's admission, therefore,created the possibility that slave states would predominate over free ones. 6s Toprevent this political shift and the policy consequences it would threaten,Northerners in Congress insisted that Maine enter the Union as a free state as aconcession for their allowing Missouri to enter as a slave state.6 6 The MissouriCompromise established a pattern that Congress would follow for some time,with slave and free states entering together, to preserve the Senate's sectionalbalance.6' This so-called balance rule "protected Northerners against thedominance of national policymaking by the South, and it protectedSoutherners against the antislavery initiatives of the North."'6 8 In other words,it enabled Congress - in a bipartisan and bisectional manner - to lock in thenational status quo on the slavery question.

But sectional balance in the Senate did not survive. By the early 186os,Republicans had control of Congress, and the party was then able to use stateadmissions policy to enact its national agenda and insulate that agenda againstsubsequent reversal by Democrats. The admission of Nevada in 1864 providesone early and stark example of this strategy. When the state was admitted in1864, Nevada's population was approximately forty thousand, and its economywas undeveloped.169 Thus, Nevada's admission was "the most egregious effortin the nation's history to disregard population and economic criteria in order to

164. See THE FREDERICK DOUGLASS ENCYCLOPEDIA 123 (Julius E. Thompson et al. eds., 2010); H.Jason Combs, Slavery in the Platte Region, 15 NEB. ANTHROPOLOGIST 8, 9 (1999).

165. POTTER, supra note 162, at 53.166. See FEHRENBACHER, supra note 162, at 107.

167. Weingast, supra note 163, at 154 tbl.4.1. The pattern predates the Missouri Compromise, andmay indeed date back to the Founding and the admission of Vermont as a counterbalance toKentucky.

168. See id. at 151.

169. See generally RUSSELL R. ELLIOTT, HISTORY OF NEVADA 70-71, 99 (1973). Nearly a centurybefore Nevada was admitted, "[w]hen the Northwest Territories were divided into states,Congress required a population of 6o,ooo for each territory to be admitted." Lawrence M.Frankel, Comment, National Representation for the District of Columbia: A Legislative Solution,139 U. PA. L. REV. 1659, 1678 (1991) (citing Northwest Ordinance of 1787, § 14, art. VI, ch. 8,1 Stat. So, 51 n.(a)).

444

125:4oo0 2015


admit a state for political reasons."' Those political reasons were clear:Nevada may well have been admitted "to bolster Republican numbers in theSenate" and to "provide votes for the ratification of the Thirteenth Amendment[and] Lincoln's reelection in 1864.""' The State did both."'

Across the 186os and 1870s, Republicans continued to admit Republicanstates to the Union, enabling the party to lock up control of the Senate andthereby enact-and entrench-its favored policies." Summarizing theRepublican Party's approach to state admission policies across the period,Charles Stewart and Barry Weingast write:

Republican political hegemony in the 186os allowed them a head startin the race to admit new states. During the secession crisis,congressional Republicans took advantage of the withdrawal ofsouthern members to admit Kansas as a free, and Republican, state.Over the objections of the few remaining Democrats, Congressaccepted the Unionist government in Wheeling as the legitimategovernment of Virginia, accepted its vote consenting to the partition ofVirginia, and admitted West Virginia as a new state. While denyingadmission to the more populous (but Democratic) Utah, Congressvoted to admit (Republican) Nevada when its population was only onefifth that of the next-smallest state and one seventh that of Utah. By thetime the South fully returned to Congress . . . one sixth of theRepublican delegation in the Senate came from states admitted duringthe Civil War and Reconstruction, and three of these four states ...provided a nearly solid core of Republican voting strength in the Senatefor the rest of the century.74

170. Charles Stewart III & Barry R. Weingast, Stacking the Senate, Changing the Nation:Republican Rotten Boroughs, Statehood Politics, and American Political Development, 6 STUD.

AM. POL. DEV. 223, 232 (1992).

171. Eric Biber, The Price of Admission: Causes, Effects, and Patterns of Conditions Imposed on StatesEntering the Union, 46 AM. J. LEGAL HIST. 119, 141 & n.76 (2004); see also ELUOir, supra note169, at 83-84; LESLIE BURNs GRAY, THE SOURCE AND THE VISION: NEVADA'S ROLE IN THE

ClVIL WAR AMENDMENTS AND THE RECONSTRUCTION LEGISLATION 31,45 (1990); GILMAN M.

OSTRANDER, NEVADA: THE GREAT ROTTEN BOROUGH 1859-1934, at 35 (1966).

172. See Stewart & Weingast, supra note 170, at 236 ("Nevada regularly sent Republicans toCongress, in both chambers, for the next thirty years. It dutifully provided three additionalelectoral votes for Lincoln in 1864, ratified the Thirteenth Amendment, and continued tovote Republican until 1876.").

173. See id. at 246, 270.

174. Id. at 227 (footnote omitted).

445


More contemporary state admissions debates also reflect theseentrenchment dynamics. When Alaska and Hawaii's admissions were beingdebated in 1953, Republican congressional majorities opposed statehood forAlaska because it was a historically Democratic territory; they supportedHawaii's admission because of that territory's more conservative politicalconstituency.17' Likewise, in the ongoing debate over statehood for the Districtof Columbia, Republicans oppose admission and Democrats support it becauseof the predictable partisan impact that D.C. statehood would have on thecongressional balance of power.176

2. From Suffrage Restrictions to Immigration

Immigration policy is another obvious lever for expanding or restricting thescope of the political community. Unsurprisingly, it too has been usedthroughout American history as a mechanism of entrenchment.'" In 1798, forexample, Congress increased the number of years an immigrant had to bepresent in the United States before naturalization from five to fourteen andthus significantly delayed the enfranchisement of newly arrived Americanresidents.'78 Although there were likely a range of motivations for thesepolicies, Adam Cox and Eric Posner argue that this delay was orchestrated bythe Federalist Party in part as a means of preventing immigrants frominstalling Jeffersonians in power.179 Daniel Tichenor similarly describes it as"an effort by the Federalist party to forestall its imminent loss of politicalpower.",so When, in the elections of 18oo, the Federalists did lose their hold onnational political power, the Democratic-Republicans who took control of thefederal government changed the residency requirement back to five years.'8

The political consequences of these shifts in immigration law were not lost on

175. See Jonathan S. Ross, Note, A New Answer for an Old Question: Should Alaska Once AgainConsider a Unicameral Legislature?, 27 AIASKAL. REV. 257, 262-63 (2010).

176. See, e.g., Peter Raven-Hansen, The Constitutionality ofD.C. Statehood, 60 GEO. WASH. L. REV.160, 161-62 (1991).

177. See Adam B. Cox & Eric A. Posner, The Rights ofMigrants: An Optimal Contract Framework,84 N.Y.U. L. REV. 1403, 1447 (2009).

178. Naturalization Act of 1798, ch. 56, 1 Star. 566 (repealed 1802). The five-year residence periodhad been established by the Naturalization Act of 1795, ch. 20, 1 Stat. 414, which wasrepealed in 1802.

179. Cox & Posner, supra note 177, at 1448.

i8o. DANIEL J. TICHENOR, DIVIDING LINES: THE POLITICS OF IMMIGRATION CONTROL IN AMERICA

54 (2oo2).

181. See Naturalization Act of 1802, ch.28, 2 Stat. 153.

125:40oo 2015


contemporary observers. Tichenor recounts how "Federalist newspapers likethe Columbia Sentinel featured naturalization policy in an extended seriesexploring how Jeffersonians translated proimmigrant policies into foreign-born votes.",,8,

In the 18oos, Democrats continued to push for liberal immigration andnaturalization policies, in part because of their predictable electoral effects. Atthe local level, "Democratic organizations worked hard to enfranchise whitemale newcomers as swiftly as possible," and, in 1845, a congressionalinvestigation found that "urban Democratic political machines were wellpracticed at naturalizing thousands of immigrants just before elections."8 3

That political dynamic continues to prevail in contemporary politics, withthe Republican Party pushing to delay the naturalization of immigrantscurrently residing in the United States, and the Democratic Party attempting toensure that naturalization. Thus, under President Clinton, the Immigrationand Naturalization Service launched a program called Citizenship USA, whichaimed to speed up naturalization.'* Republicans at the time viewedCitizenship USA as an attempt to increase the number of Democratic voters,and charged that a top aide to Vice President Gore had called the program a"pro-Democratic voter mill."""s

Similarly, the immigration bill passed by the Democratic-controlled Senatein June 2013 contained a path to citizenship for the eleven millionundocumented immigrants currently living in the country.'8 6 This populationconsists primarily of low-income Hispanics, who vote by great margins forDemocratic candidates .8' The undocumented population, if naturalized,would also constitute a significant percentage of the electorate in several statescrucial to presidential politics: by 2020, formerly undocumented immigrantsare projected to make up 7.1% of the electorate in Texas, 6.7% in Arizona, 7.8%

182. TICHENOR, supra note 18o, at 55.

183. Id. at 59.

184. Sara Fritz, Gore Immigrant Program Role Draws Fire, L.A. TIMEs (Oct. 6, 1996), http://articles.latimes.com/1996-lo-o6/news/mn-51224-1_vice-president [http://perma.cc/8MYP-VR62].

185. Id.

186. S. 744, 113th Cong. (2013).

187. See Mark Hugo Lopez & Paul Taylor, Latino Voters in the 2012 Election, PEW RES. CTR. (Nov.

7, 2012), http://www.pewhispanic.org/2012/1/o7Aatino-voters-in-the-2012-election [http://perma.cc/S9KQ-BAKY]; Jeffrey S. Passell & D'Vera Cohn, A Portrait of UnauthorizedImmigrants in the United States, PEW RES. CTR. (Apr. 14, 2009), http://www.pewhispanic.org/2009/o4/14/a-portrait-of-unauthorized-immigrants-in-the-united-states [http://perma.cc/Z4JX-8F82]; A Nation of Immigrants, PEW RES. CTR. (Jan. 29, 2013), http://www.pewhispanic.org/2013/01/29/a-nation-of-immigrants [http://perma.cc/DRK2-7NUS].

447


in Nevada, 4.9% in Florida and Georgia, and between 2% and 4% in Colorado,Virginia, and North Carolina.'88

While Democrats supported citizenship for these undocumented residents,Republicans in the Senate not only opposed the path to citizenship, but alsooffered amendments that would have lengthened the time to naturalization andmade the requirements for naturalization more onerous than they are today.189Republican commentators are not shy about defending this decision onelectoral grounds. Thus, Laura Ingraham argued in a Washington Post opinionpiece that the GOP should continue to oppose immigration reform: "In light of. . . the experience of California-which has shifted from a Republicanstronghold to one of the most liberal states in the country, in large part becauseof the rise of its immigrant population-it is absurd to pretend that allowingeven more immigrant voters wouldn't be a boon to the Democrats."'9 oSimilarly, Rush Limbaugh called immigration reform "Republican suicide."

D. Switching Decision Makers

The previous two entrenchment strategies operate by shifting the relativepower of groups with a say in the political decision-making process. A furtherstrategy is to shift the locus of political decision making, empowering adifferent set of political actors and groups. For example, the delegation ofauthority to independent central banks is often viewed as a mechanism forresisting political demands by short-sighted politicians and popular majoritiesfor inflationary and otherwise misguided monetary policies.'92 Central banksthat can be successfully insulated from these political pressures enable

188. Walter Hickey, How a Path to Citizenship for Illegal Immigrants Would Influence thePresidential Election of2o2o, Bus. INSIDER (Mar. 20, 2013), http://www.businessinsider.com/impact-of-a-path-to-citizenship-on-2020-election-2013-3 [http://perma.cc/6U9-5HGZ].

18g. See, for example, Senator Jeff Sessions's amendments in committee markup and on thefloor. For a full summary of amendments considered during committee markup, see S. REP.No. 113-40 passim (2013), http://www.congress.gov/13/crpt/srpt4o/CRPT-I13srpt4o.pdf[http://perma.cc/5RK8-RYGP].

190. Laura Ingraham, Opinion, Why Conservatives Should Say No to Immigration Reform, WASH.PosT (Feb. 19, 2014), http://www.washingtonpost.com/opinions/laura-ingraham-why-conservatives-should-say-no-to-immigration-reform/2o14/o2/19/85ae3438-98fo-n1e3-b

9 31-0204122c514b-story.html [http://perma.cc/W2R4-2JC4].

191. The Amnesty Bill Is GOP Suicide, RusH LIMBAUGH SHOW (Apr. 26, 2013), http://www.rushlimbaugh.com/daily/2013/o4/26/the-amnesty-bill-is-gop-suicide [http://perma.cc/KQ29-VR9K].

192. See ALAN DRAZEN, PoLTIcAL ECONOMY IN MACROECONOMICS 144 (2000); Kenneth Rogoff,The Optimal Degree of Commitment to an Intermediate Monetary Target, ioo QJ.. ECoN. n69(1985).

125:4oo 2015


governments to entrench sound monetary policies. Independent central banksmay also enable "political leaders [to] bind the hands of their successors in theformation of monetary policy," locking in their preferred policies even afterthey have been voted out of power.'

As this example illustrates, relocating decision-making authority toinstitutional actors that are relatively insulated from political forces can shieldpolicies from change. Variations on this strategy, discussed below, includedelegations to courts, administrative agencies, and international governancebodies. To the extent these institutions are likely to be controlled by politicalallies of the delegator, and to the extent the delegation will be relativelydifficult to retract, 94 this can be an effective mechanism of policyentrenchment.

1. From Legislative Entrenchment to Judicial Entrenchment

Social scientists and legal scholars alike have recognized that politicians canentrench their policies and protect their hold on power by delegating decision-making authority to a politically insulated judiciary. Thus, working within apublic choice paradigm, William Landes and Richard Posner famously arguedthat an independent judiciary was a useful tool for legislators who wanted todeliver statutory benefits for interest groups in exchange for campaigncontributions and political support. By enforcing the initial bargains againstdownstream legislatures with political incentives to renege, Landes and Posnerargued, courts could increase the durability-and thus the value-of theseinterest group bargains.919

In the United States, since the early days of the Republic, presidents andparties often have resorted to a similar strategy of "political entrenchment inthe judiciary" to preserve their preferred policies in the face of political defeat.Having lost control of the national government to the Republicans in theelection of 18oo, the lame-duck Federalist Congress famously passed the 18oJudiciary Act, the so-called "Midnight Judges Act," expanding the size and

jurisdiction of the federal judiciary and creating the opportunity for PresidentAdams to appoint a number of loyal Federalists to life terms on the bench.Furious that the Federalists had packed the judiciary on their way out of office,newly elected President Jefferson clearly understood the political strategy in

193. John B. Goodman, The Politics ofCentral Bank Independence, 23 COMP. POL. 329, 334 (1991).

194. This condition is crucial, yet often overlooked. See Levinson, supra note 17, at 681-83.

195. See William M. Landes & Richard A. Posner, The Independent Judiciary in an Interest-Group

Perspective, 18 J.L. & EcoN. 875 (1975).

449


play: "[The Federalists] have retired into the judiciary as a stronghold . .. andfrom that battery all the works of republicanism are to be beaten down anderased. "96

This strategy was executed more effectively by Republicans in the latenineteenth century. As they perceived the electoral tide beginning to turnagainst them, the Republican Party sought to lock in their agenda ofconservative "economic nationalism" by expanding the jurisdiction of thefederal courts and staffing them with ideologically sympathetic judges andJustices.197 As the political scientist Howard Gillman has described, the"increased power, jurisdiction, and conservatism of federal courts during thisperiod was a by-product of Republican Party efforts to promote and entrench apolicy of economic nationalism during a time when that agenda was vulnerableto electoral politics."98

Jack Balkin and Sanford Levinson generalize from examples like this toemphasize the importance to American political and constitutionaldevelopment of what they call "partisan entrenchment": the strategicappointment of politically and ideologically aligned judges and Justices whosetenures will outlast party control over the political branches of government.199Balkin and Levinson emphasize that presidents and parties are motivated toengage in partisan entrenchment not just to "secure a bench likely to assist thePresident with his current political agenda" but also "to secure future influenceeven when the party loses power.""o From this perspective, federal judges andJustices are simply "temporally extended representatives of particular parties,"or representatives of "a temporally extended majority," and "hence, of popularunderstandings about public policy and the Constitution."" Thus, picking upshortly after the historical point where Gillman leaves off, Balkin and Levinsonpoint to the early New Deal period, in which "the federal judiciary, which hadbeen entrenched by the Republican Party, mostly resisted the Democrats'

196. Letter from Thomas Jefferson to John Dickinson (Dec. 19, 18oi), in lo THE WRITINGS OFTHOMAS JEFFERSON 302, 302 (Andrew A. Lipscomb & Albert Ellery Bergh eds., 1904).

197. Howard Gillman, How Political Parties Can Use the Courts To Advance Their Agendas: FederalCourts in the United States, 1875-1891, 96 AM. POL. Sci. REv. 511, 512-13, 516-17 (2002).

198. Id. at 511.

199. Jack M. Balkin & Sanford Levinson, Understanding the Constitutional Revolution, 87 VA. L.

REV. 1045, io66 (2001).

200. Jack M. Balkin & Sanford Levinson, The Processes of Constitutional Change: From PartisanEntrenchment to the National Surveillance State, 75 FoRDHAM L. REv. 489, 495 (20o6).

zoi. Balkin & Levinson, supra note 199, at 1o67, 1076.

450

125:4oo0 2015


proposed construction of the regulatory and welfare state.""2 Examples likethis could be extended through the present. For instance, one account of thepolitics of judicial appointments during Ronald Reagan's presidencyemphasizes that the administration "approached [judicial] appointments as away 'to institutionalize the Reagan revolution so it can't be set aside no matterwhat happens in future presidential elections."'3

2. From Legislative Entrenchment to Administrative and InternationalEntrenchment

A close analogue to judicial entrenchment is administrative entrenchment.Temporarily prevailing parties and political coalitions can extend theirinfluence beyond the boundaries of political defeat by delegating decision-making authority to an administrative agency that is relatively insulated frompolitical control, effectively "lock[ing] in policies so they are not reversed orundone when political power changes hands.""4 For example, McNollgastportrays the 1946 enactment of the Administrative Procedure Act (APA) as astrategy for entrenching the New Deal.2 os When it became increasingly likelythat the Democrats would lose the White House in 1948, the Party-while stillin Congress - supported the imposition of procedural restrictions on theadministrative agencies it had recently created. According to McNollgast, theDemocrats supported the APA and its procedural restraints on agency action as

202. Balkin & Levinson, supra note 200, at 534. From a normative standpoint, Balkin andLevinson for the most part view partisan entrenchment through the judiciary as an attractivefeature of American democracy, one that plays the important role of mediating the tensionbetween constitutionalism and democratic self-government. Id. But they also believe it cansometimes go too far. Thus, when it comes to the Supreme Court's decision in Bush v. Gore,

531 U.S. 98 (2000), Balkin and Levinson condemn what they see as "five members of theCourt using their powers of judicial review to entrench their party in the Presidency, andthus, in effect, in the judiciary as well, because of the President's appointments power."Balkin & Levinson, supra note 199, at 1o8o. Why we should be fine with Presidents andparties entrenching their policy positions through judicial appointments, but not withPresidents entrenching themselves in office by way of favorable rulings about election lawcoming from their judicial appointees, is not entirely clear.

203. Reva B. Siegel, Dead or Alive: Originalism as Popular Constitutionalism in Heller, 122 HARV. L.REv. 191, 220 (2008) (quoting David M. O'Brien, Opinion, Meese's Agenda for Ensuring theReagan Legacy, L.A. TIMES (Sept. 28, 1986), http://articles.latimes.com/1986-o9-28/opinion

/op-9537-_iedwin-meese-iii [http://perma.cc/K3U7-3RAF]).

204. Matthew C. Stephenson, Statutory Interpretation by Agencies, in RESEARCH HANDBOOK ONPUBLIC CHOICE AND PUBLIC LAW 285, 288-89 (Daniel A. Farber & Anne Joseph O'Connelleds., 2010) (surveying the relevant literature); see also supra note 13 and accompanying text.

2os. McNollgast, The Political Origins of the Administrative Procedure Act, 15 J.L. ECON. & ORG.18o, 180-83 (1999).

451


a means of "'hard wir[ing]' the policies of the New Deal against an expectedRepublican, anti-New Deal political tide." " 6 In particular,

[t]he danger was that a Republican president could use this broaddiscretion to undo much of New Deal regulatory policy simply byappointing anti-New Dealers to head these agencies. Since proceduralrestraints make it costly and politically difficult for agencies to changeexisting policy, the establishment of procedural due process wouldblunt any Republican president's ability to dismantle or shift theregulatory policies of the New Deal."

Decisions about administrative agency design can also entrench policy bysetting into motion the type of policy-mobilization cycle described earlier. AsJonathan Macey argues, the design of administrative agencies can "perpetuatethe power and legitimacy of certain [interest] groups and undermine the powerand legitimacy of others."2os Interest groups empowered by a particular agencydesign will, of course, be the groups most likely to influence agency decisionmaking in the future. So, legislatures can entrench their policy preferences bydesigning agencies in ways that empower the interest groups that share thepreferences of the enacting legislature.7"

The executive can use agencies as an entrenchment device too. One way isthrough what Nina Mendelson has called "agency burrowing," which occurswhen an outgoing, lame-duck presidential administration engages in last-minute attempts to entrench its policy preferences through various executive

2o6. Id. at 18o.207. Id. at 192. Once again, we might question why the Republicans did not just enact a statute

repealing or revising the APA. The possibility of entrenchment by way of delegation to, ordesign of the structure and process of, administrative agencies assumes that thesearrangements cannot be disentrenched as easily as they were enacted. See Stephenson, supranote 204.

2o8. See Jonathan R. Macey, Organizational Design and Political Control ofAdministrative Agencies,8 J.L. EcoN. & ORG. 93,99 (1992).

209. As one example of this design dynamic, Macey points to the decision about whether anagency will regulate a range of industries with distinct interests (the Occupational Safetyand Health Administration, for example) or whether the agency will be structured so as toregulate a single industry (the Commodities Futures Trading Commission, for example).See id. at 99, 104-08. Macey's argument is that an agency with a single-industry jurisdictionis more likely to empower that industry's interest group than is an agency that regulates arange of industries and thus competing interest groups. See id. at 104-08. The structure anddesign of an agency can ensure that the policy preferences of the regulated industry willcontinue to be enacted by the agency, even if the electoral majority that established theagency is no longer dominant. See id. at 108-09.

452

125:4oo0 2015


actions.'o On the model of partisan entrenchment in the judiciary, theadministration can engage in "personnel burrowing," converting agencypolitical appointees into civil service employees in order to keep them in theirjobs beyond the end of the President's term."' The idea is that these appointeeswill carry the (former) President's policy preferences forward with them."

Along similar lines, Elizabeth Magill has described how administrativeagencies often impose limitations on their own discretion to act as a means ofensuring that policies implemented by the sitting agency are difficult to changeby the appointees of the subsequent administration. For example, agenciessometimes offer greater procedural protections than the APA requires." Just asprocedural rules imposed by the APA create hurdles to policy change, this andother forms of agency "self-regulation" can increase these hurdles and morefully insulate the policy status quo. As Magill elaborates, a "self-regulatorymeasure might create a process that involves so many key actors that the statusquo bias would be great because it takes so many to agree to change policies orbecause the specific actors empowered under the regime will predictably holdparticular views."s

The same basic strategy of entrenchment can be pursued through"upward" delegations to international organizations. Scholars of internationalrelations have described how temporarily prevailing political coalitions canmake use of international agreements to preserve policies threatened bydomestic politics by placing them under the control of internationalorganizations that are insulated by a democracy deficit." 6 Thus, looking at thehuman rights context in the late 1940s, Andrew Moravcsik observed thatreciprocally binding human rights obligations tended to be supported only bynewly established - and therefore tenuous - democratic governments, and that

21o. Nina A. Mendelson, Agency Burrowing: Entrenching Policies and Personnel Before a New

President Arrives, 78 N.Y.U. L. REV. 557, 560-61 (2003).

211. Id. at 606.

212. Id. at 6o8.

213. Elizabeth Magill, Agency Self-Regulation, 77 GEO. WASH. L. REv. 859, 861-62 (2009).

214. Id. at 868.

215. Id. at 888. Such self-regulatory agency behavior is particularly effective given the judicial

doctrine that requires agencies to adhere to their own procedural rules. See id. at 873-76.

216. See, e.g., Frederick M. Abbott, NAFTA and the Legalization of World Politics: A Case Study, 54

INT'L ORG. 519 (2000); Rachel Brewster, The Domestic Origins ofInternational Agreements, 44

VA. J. INT'L L. 501, 511-24 (2004); Tom Ginsburg, Locking in Democracy: Constitutions,Commitment, and International Law, 38 N.Y.U. J. INT'L L. & POL. 707, 727-34 (2oo6); Beth A.

Simmons & Allison Danner, Credible Commitments and the International Criminal Court, 64INT'LORG. 225, 231-36 (2oo).

453


they tended to be rejected by both established democracies and dictatorships . 17

Moravcsik explains this phenomenon with an entrenchment analysis: thecurrent political leadership of new democracies signed on to binding humanrights agreements in order to "'lock[] in' the domestic political status quoagainst their nondemocratic opponents."`

Even better, both types of delegation-to domestic agencies and tointernational decision-making processes -can be combined to create a kind ofdouble entrenchment. This is the story told by Rachel Brewster about theentrenchment of banking regulation. 9 A group of central banks, including theU.S. Federal Reserve, agreed through the Basel Accords to establish minimumcapital requirements for banks.2 "o From the perspective of the U.S. politicalprocess, the requirements contained in the Accords might be viewed as doublyentrenched: set as an initial matter by the Federal Reserve, an insulatedpolitical actor, and then entrenched against change by the Federal Reserve itselfby the greater political costs of violating international agreements.

E. Summary

Political entrenchment can occur formally, but it can also occurfunctionally. Incumbent power holders can preserve their hold on office bymanipulating the formal rules of election law; but they can also manipulatesubstantive policy in order to neutralize their political opposition, mobilizetheir own supporters, and define the polity that will decide their next election.Likewise, if formal legislative entrenchment is off the table, power holders canlock in their policies and programs by organizing and empowering aconstituency that will resist retrenchment or by delegating decision-makingauthority to an institution that will remain politically committed topreservation.

A recent article by Richard Lazarus, tellingly entitled Super Wicked Problemsand Climate Change: Restraining the Present To Liberate the Future, offers a usefulillustration of the discussion so far." In a moment of hope that Congressand the President would enact climate change legislation, Lazarus takesas his project to design this potential statute in such a way as to entrench

217. Andrew Moravcsik, The Origins of Human Rights Regimes: Democratic Delegation in PostwarEurope, 54 INT'L ORG. 217, 219-20 (2000).

218. Id. at 244.

219. Brewster, supra note 216, at 517-18.

220. Id. at 517.

221. Lazarus, supra note 14.

454

125:4oo 2015


it against political opposition in the future. He recommends, andsketches, "precommitment strategies" that would "insulate programmaticimplementation to a significant extent from [the] powerful political andeconomic interests" that will predictably be opposed to, and intentupon eroding, an aggressive environmental regulatory regime.m"[N]otwithstanding th[e] undemocratic effects" of such strategies, Lazarusbelieves that current lawmakers would be justified in "making it more difficultfor future legislators and agency officials to substitute their views of soundpolicy for the judgment of past lawmakers" for the purpose of saving the planetfrom the catastrophic effects of climate change.

Of course, one way of entrenching a climate change statute would be tomake it formally unrepealable. Lazarus considers this option, but ultimatelyrejects it, in part because legislative entrenchment is of dubiousconstitutionality and would become a source of controversy, and in partbecause an absolute ban on amendment would be too extreme.'

Instead, Lazarus recommends a series of functional entrenchmentstrategies. One possibility would be to "design federal climate legislation in amanner that would create a powerful political constituency with a strongeconomic incentive favoring the legislation's preservation" -for example, byincluding a tradable emissions program that would attract large investments inemissions rights.22 Another would be "to couple domestic climate changelegislation with the United States' agreement to international treatyobligations," which would "significantly raise the political cost of anyretreat."'6 Other measures would be designed to protect and promote climatechange regulation in the executive branch, pursuant to statutory delegation.On the model of the Federal Reserve, Lazarus suggests ways of insulatingagency officials in charge of implementation from "pressures likely to derivefrom short-term economic concerns, which [might] undermine the law'seffectiveness."7̀ He also suggests ways to structure the regulatory process inorder "to enhance the influence of interest groups that are concerned aboutprotecting future generations but which otherwise lack the necessary economicor political clout.""5 As Lazarus clearly appreciates, these strategies of

222. Id. at 1158.

223. Id. at 1157.

224. Id. at 1207-09.

225. Id. at 1210.

226. Id. at 1209.

227. Id. at 1212; see also id. at 1212-16 (describing strategies for insulating agency officials).

228. Id. at 1212; see also id. at 1216-25 (describing strategies for enhancing interest-group clout).

455


functional entrenchment serve as close substitutes for formal entrenchment ofeither the legislative or electoral varieties.

The Bush tax cuts of 2001 and 2003 provide another vivid illustration of thesubstitutability of formal and functional entrenchment. The tax cuts wereenacted with sunset provisions, so that they would automatically expire in2010.229 This raised no flags from a legal perspective; quite the contrary, sunsetprovisions are precisely the opposite of the kind of formal legislativeentrenchment of which public law disapproves. Yet political observers noticedthat the tax cuts, even while couched as temporary for purposes of facilitatingenactment and disguising their likely budgetary implications, were in factengineered to be self-entrenching.2 30 The reforms were "tailored to shape thepolitics of tax cuts down the line in ways that favored tax-cutters' long termgoals,"231 creating a political dynamic that would lead to their extension-essentially duplicating the effect of an unrepealable statute. Furthermore,linking legislative to electoral entrenchment, Republicans might well haveexpected that political support for the tax cuts would "provide a powerfulmotivation for the wealthy to bankroll Republican reelection efforts in thefuture."23 ' Once again, the moral is that, even if the pathways of formalentrenchment are closed, there will often be a functional pathway that leads tothe same destination.

III. RATIONALIZING ENTRENCHMENT?

We have seen that courts and public law scholars view formal electoral andlegislative entrenchment as matters of grave normative concern and asappropriate targets for legal regulation. We have also seen that in manycontexts functional entrenchment strategies appear to be close substitutes forthese formal ones. Yet the seemingly ubiquitous phenomenon of functionalentrenchment has never been perceived as problematic in public law.23' And to

229. For a description of the tax cuts and their sunset provisions, see Rebecca M. Kysar, The SunAlso Rises: The Political Economy of Sunset Provisions in the Tax Code, 40 GA. L. REV. 335, 338,370-83 (2006).

230. JACOB S. HACKER & PAUL PIERSON, OFF CENTER: THE REPUBLICAN REVOLUTION AND THE

EROSION OF AMERICAN DEMOCRACY 60-62 (2005) (describing this feature and quoting onecommentator's observation that "[t]he lesson of the Bush tax-cutting record is that whatmatters is structural change and political leverage down the line").

231. Id. at 58.

232. Id. at 61.

233. This descriptive claim is mostly an inference from omission. However, in response to Posnerand Vermeule's example of bicycles in the park, see supra note 103 and accompanying text,

456

125:4oo0 2015


make matters more confusing, in at least one major area-constitutional law-entrenchment is widely embraced.

In this Part, we explore whether there is any way of rationalizing publiclaw's seemingly inconsistent treatment of the various forms of politicalentrenchment. Section ILA asks whether there is any good reason for viewingformal entrenchment as a bigger problem or a more sensible target of legalregulation than functional entrenchment. Section III.B then expands the frameof analysis, first, to ask whether there is any good reason for regardingsubconstitutional entrenchment as any more problematic than constitutionalentrenchment; and second, to question whether a meaningful category ofentrenchment-or the reasons for worrying about it-can be coherentlybounded at all.

A. The Uneasy Case for Policing Formal but Not Functional Entrenchment

Is there any good reason for believing that formal entrenchment should beprohibited and policed, while functional entrenchment should be tolerated orignored? One possibility is that formal entrenchment is more harmful thanfunctional entrenchment. Another is that, even if the two forms are equallyharmful, formal entrenchment is easier to identify and therefore a moreworkable target for legal regulation. We discuss each of these possibilities inturn.

i. Harmfulness

The near-consensus view in the public law literature is that formal electoraland legislative entrenchment are socially harmful and legally undesirable, if notoutright prohibited.34 As we have seen, commentators emphasize the unfairaggrandizement of political power by upstream decision makers at the expenseof their successors, the frustration of present majority will, and the costs oflocking in bad or anachronistic policies.235 Nightmare scenarios are front and

critics of legislative entrenchment have been explicit in insisting that formal entrenchment isa different and more problematic phenomenon. See Dana & Koniak, supra note 67, at 530-31;Roberts & Chemerinsky, supra note 77, at 1813-19; Stewart E. Sterk, Retrenchment ofEntrenchment, 71 GEO. WASH. L. REv. 231, 238-39 (2003).

234. Posner and Vermeule are a major exception with respect to legislative entrenchment, andmuch of what they say in defense of that practice applies to entrenchment more broadly. SeePosner & Vermeule, supra note 24. Eule goes out of his way to recognize some of thefunctional benefits of entrenchment before proceeding to conclude that legislativeentrenchment is nonetheless unconstitutional. See Eule, supra note 67, at 390-91.

235. See supra Section I.B.

457


center. A lame-duck Democratic Congress, facing a newly elected RepublicanCongress and President, enacts a statute entrenching its entire legislativeprogram against modification or repeal.3 After the enactment of anentrenched statute defunding all nuclear weapons for ten years, an unforeseenthreat to U.S. national security arises that only nuclear weapons could deter.2"When a temporarily dominant political party or coalition takes measures tocement itself in power or block the channels of political change, the specter ofoligarchy or dictatorship menacingly looms.

The social science literature on functional entrenchment takes a morebalanced normative perspective. While social scientists certainly have notmissed the fact that self-serving political actors can use various entrenchmenttechniques to lock in their political gains at the expense of opponents and thepublic at large, the literature on functional entrenchment recognizes that, in atleast some contexts, "entrenchment is a legitimate goal in a democratic

polity.",238 After all, the risk of locking in ill-motivated or mistaken policiesmust be weighed against the possibility of insulating good policies against ill-motivated or mistaken reversals. If, for instance, preventing global warmingwill improve social welfare in the long run, then society might well benefit byinsulating these policies against short-term or partial interests as Lazaruscontends.'9 Delegation to central banks, international organizations, andindependent judiciaries can be a vehicle for entrenching sound monetarypolicies, free trade, and human rights.2o

Even if there is no reason to expect prior political decisions to besubstantively better than later ones, social scientists emphasize that stabilityand predictability can be valuable in their own right." Thus, many of thefoundational contributions to the political science and economics literature onentrenchment emphasize the benefits of credible commitment.242 Agovernment that can credibly and successfully commit itself to repaying debtsor to preserving economic entitlements will be able to borrow money on morefavorable terms or encourage private sector investment; a government that cancredibly commit not to bail out banks that make risky investments may be able

236. See Sterk, supra note 233, at 237.

237. See Eule, supra note 67, at 386-87.238. Patashnik & Zelizer, supra note 12, at 1083.239. See Lazarus, supra note 14.

24o. See supra Section II.D.

241. See Levinson, supra note 17, at 673-75 (surveying beneficial forms of commitment drawnfrom the social science literature).

242. See, e.g., id. at 673-74.

458

125:4oo 2015


to avert a future financial crisis; and so on. Entrenchment-induced stability canalso serve to "tak[e] particularly contentious issues or subjects off the table"and to allow controversies to be settled in a lasting way, economizing onconflict, reducing rent-seeking, and freeing up political resources for otheruses.4 3

In short, entrenchment can be used for good as well as for ill. But- andhere is the crucial point - this goes for formal as well as functionalentrenchment. An unamendable statutory formula could replace the delegatedauthority of the Federal Reserve as a means of locking in sound monetarypolicy. If stringent environmental regulation will provide long-term benefits inpreventing climate change, those benefits could also be achieved throughlegislative entrenchment, or perhaps alternatively by tilting the electoralplaying field in favor of liberal Democrats.

In sum, there is nothing about the distinction between formal andfunctional entrenchment that would appear to correlate with social harms orbenefits. Both can be motivated by the narrow political self-interest of parties,politicians, and interest groups; or, alternatively, by broader, public-regardingmotivations. And both can be used to accomplish the same outcomes,insulating officeholders and policies against downstream majorities withdifferent preferences. Given the broad substitutability of formal and functionalentrenchment mechanisms, there is no reason to believe that one would becategorically more harmful (or beneficial) than the other.

2. Identifiability

Despite their general substitutability, formal entrenchment does appear todiffer from functional entrenchment in one important way: it will almostalways be easier to identify. On the (questionable) assumption thatentrenchment is generally a bad thing, this could explain why formal

243. Id. at 675. While these benefits follow most naturally from the stability of policy decisions,the advantages of stabilizing electoral outcomes should be broadly similar in kind. Theentrenchment of incumbents may contribute to more effective governance, on the theorythat more experienced legislators with accumulated knowledge of their constituents andtheir needs will outperform a rotating cast of novices. The entrenchment of parties orcoalitions might also lengthen political time-horizons, facilitating the implementation ofpolicies with longer-term social benefits as opposed to focusing politicians on short-termpolitical gains. Electoral entrenchment should also reduce the frequency of dramatic,destabilizing shifts in policies that can come with changes in party control of government.And, of course, if the entrenchment of parties and political coalitions leads to theentrenchment of a platform or set of policy outcomes, then all of the potential advantages oflegislative entrenchment with respect to stability, predictability, and commitment shouldcarry over.

459


entrenchment alone is singled out by public law. If strategies of formalentrenchment can be more readily identified and distinguished from benignpolitical behavior, that could make them more sensible targets for judicialscrutiny or other forms of legal regulation than their functional brethren. 4

Whatever else might be said for the prohibition on formal legislativeentrenchment, it is simple enough to administer. Statutes that explicitlyannounce their own entrenchment through prohibitions or special proceduralobstacles to repeal are easy to identify and to distinguish from ordinarystatutes. To be sure, there is considerable ambiguity about how far beyond thiscore case the prohibition on legislative entrenchment might be extended -toframework statutes, internal legislative rules like the Senate filibuster, (sometypes of) government contracts, and so on."s But at least in the core case, thedefinitional boundaries of entrenchment are clear, and the statutes that fit thedefinition announce themselves unambiguously.

When we move beyond the simple case of legislative entrenchment,however, the task of defining and identifying a forbidden category of evenformal entrenchment becomes more difficult. Electoral entrenchment isillustrative. Despite the calls of election law scholars for a more aggressivejudicial role in policing entrenchment in this domain, courts have beendaunted by the difficulty of demarcating a judicially administrable category offorbidden conduct.4 7

One source of difficulty lies in determining the relevant baseline foridentifying impermissible entrenchment. As we have seen, diagnoses ofentrenchment typically rely upon baselines set by vague reference to the"ordinary" or ideal difficulty of effecting political change. 2' In the context ofelection law, courts and commentators have struggled over what, precisely, theright baseline should be.49 When it comes to the partisan gerrymandering ofelection districts, for example, political entrenchment must be measuredagainst some baseline of politically "fair" districting. What the metric of

24. Cf McGinnis & Rappaport, supra note 24, at 442 (arguing that there are "strong reasonsgrounded in administrative costs" for policing formal entrenchment mechanisms but not"informal ones," such as Posner and Vermeule's bicycles in the park example).

245. We return to these ambiguities below. See infra Section III.B.2.

246. See supra notes 60-64 and accompanying text.

247. See supra note 5o and accompanying text.

z48. See supra Section I.A.

249. See Klarman, supra note 4, at 533-34. As Klarman (among others) has argued, these baselineproblems can be sidestepped by a procedural solution to the problem of entrenchmentthrough gerrymandering- turning over districting to some sort of impartial commission orto a computer program. See id. at 534.

46o

125:4oo0 2015


fairness should be, however, is not clear. Courts and commentators havedisagreed for decades about whether an unfair, or entrenching, partisandistribution of districts should be determined by reference to a baseline ofmajority rule, proportional representation, or something else.25o In Vieth v.Jubelirer, a plurality of the Supreme Court threw up its collective hands,describing the attempt to arrive at a judicially manageable standard as"[e]ighteen years of judicial effort with virtually nothing to show for it.""

A further difficulty is that laws that may have the purpose and effect ofentrenching a party or policy may also have other purposes and effects worthyof democratic respect. Here again, election law is illustrative.s2 Campaignfinance regulations might well benefit incumbents at the expense of challengersor one party at the expense of another, but they may also aim to mitigate theinfluence of wealth on elections and equalize the political influence ofconstituents across income groups."' Ballot access restrictions and voteridentification requirements can be used to entrench parties and incumbents,but they can also play a legitimate role in preventing voter confusion andfraud.4 Multimember districts can be used to dilute the voting power of racialminority groups and entrench white majorities and their preferredrepresentatives, but they can also generate representatives who are responsiveto the needs of the broader political community rather than to local, parochialinterests.2 ss How courts would police the purpose or effects of laws like this toscreen for entrenchment, and how they would balance the benefits ofpreventing entrenchment against the costs of sacrificing the non-entrenchment-related benefits of such laws, is also unclear.

Whatever the prospects for overcoming these dual difficulties ofindeterminate baselines and mixed motives in the election law context,"'extending antientrenchment review to the vast universe of functionalentrenchment would seem to present a challenge of a different order.

Starting with baselines, any attempt to operationalize a prohibition on thefunctional entrenchment of parties or political coalitions would have to

250. See id. at 533-34.

251. 541 U.S. 267, 281 (2004) (plurality opinion); see also Pildes, supra note 42, at 66-83(discussing the difficulty of formulating a standard for partisan entrenchment).

252. See Klarman, supra note 4, at 529-30 (describing the problem of mixed motives in the

context of election law).

253. See id. at 536-38.

254. See id. at 535-36.

255. See id. at 538.

256. See id. at 528-39 (suggesting ways in which courts might successfully implement a regime of

antientrenchment review in the context of election law).

461


confront the fact that many changes in policy might be viewed as tilting thepolitical playing field in favor of one side or the other. Imagine the position of acourt charged with policing against partisan entrenchment through labor law.On the one hand, an expansive and protective labor statute might be viewed asentrenching Democrats because it facilitates the growth of unions, which inturn support Democratic candidates. On the other hand, a restrictive andpunitive labor statute might be viewed as entrenching Republicans. From theperspective of a status quo ante baseline, any change in labor law could beviewed as entrenchment because one party or the other will likely benefitpolitically. Even no change in labor law might be viewed as entrenching, if thedecision to do nothing were cast as refusing or failing to enact some salientpro- or anti-labor policy. The same quandary would arise in assessing anychange (or even the absence of any change) in policy in the domains of tortreform, immigration, and many other areas with predictable political feedbackeffects.

As we have seen, it is not even clear whether-or when-the status quoante should be the dispositive baseline for measuring entrenchment.5 7 If welooked instead to majoritarian preferences, that might lead to a different viewof partisan entrenchment. Returning to the labor law example, suppose aprounion policy shift has the effect of bringing previously disfranchisedvoters to the polls and thereby pulling partisan and policy support closerto majoritarian political preferences. Now a prior enactment diagnosisof entrenchment competes with a majoritarian baseline diagnosis ofdis-entrenchment. A court called upon to police labor law for partisanentrenchment might well wonder what to do.

Comparable problems would arise in policing the entrenchment of anypolicy. In the absence of formal criteria for entrenchment, some assessmentwould have to be made of the difficulty of revising or repealing the policy onceenacted, as compared to some baseline measure of the "ordinary" difficulty ofpolicy change. A countermajoritarian measure seems unpromising, given thepervasively countermajoritarian tendencies of the American political system.sAccording to one recent study, the probability that a policy change supportedby three-quarters of Americans will be enacted into law is only thirty-ninepercent.2 59 And, as elaborated below, it is quite likely that many, perhaps evenmost, of the policies that have been enacted into law would not have beenchosen by present majorities. Short of concluding that much of current law is


258. See infra notes 287-288 and accompanying text.

259. See MARTIN GILENs, AFFLUENCE AND INFLUENCE: ECONOMIC INEQUALITY AND POLITICAL

POWER IN AMERICA 74 (2012).

462

125:40oo 2015


impermissibly entrenched, it is hard to know how courts (or other assessors)would proceed.

Switching to a status quo ante baseline would raise a different set ofchallenges. The question would then become whether a policy was, or waslikely to become, more politically difficult to revise or repeal than it had been toenact in the first place. Answering this question would require some empiricalassessment of the difficulty of effecting political change at different points intime. Such an assessment would need to control for the political popularity ofthe relevant policy. Again, entrenchment entails some impediment to therealization of political preferences; the accurate reflection of increasinglyfavorable preferences would be evidence of democratic responsiveness, theopposite of entrenchment. But distinguishing political popularity from thekinds of "artificial" impediments to policy change that should qualify asentrenchment is not straightforward. When should we view the strengtheningof political opposition to change as a structural barrier evincing entrenchmentas opposed to a perfectly valid expression of democratic will? Should we alwaysthink of policies that mobilize and empower supportive interest groups asbecoming entrenched, or are there conditions under which we should view thisdynamic as legitimately increasing political support? Answering thesequestions will require a much fuller account of the difference between well-functioning and distorted democracy than existing theories of entrenchmenthave contemplated.o

The problem of mixed motives and effects also seems more pervasive andsevere as applied to functional entrenchment. As we have seen, labor law, tortreform, and immigration policy can serve as entrenchment mechanisms; butthey are also labor law and tort reform and immigration policy. Whenentrenchment is accomplished through changes in substantive policy, therewill always be a plausible case to be made that the motivation was policy ratherthan politics, and, to the extent this is true, invalidating legislation onentrenchment grounds would mean blocking the enactment of democraticallypreferred policies. Something similar is true of functional entrenchment byway of changing the locus of political decision making. Decisions to delegate tocourts, administrative agencies, and international governance bodies createpotential benefits that stand entirely apart from entrenchment and that wouldbe sacrificed if these decisions were blocked on anti-entrenchment grounds.Whatever the democratic costs of entrenchment, the democratic costs ofpreventing entrenchment might also be substantial.

26o. As we discuss further below, once we let go of formal markers -explicit prohibitions onrepeal, supermajority requirements, and the like-the precise definition of entrenchmentquickly begins to blur. See infra Section III.B.2.

463


All of these conceptual difficulties in identifying functional entrenchmentwould be compounded by empirical problems. Courts charged with policingentrenchment presumably would need to predict the effects of policyenactments and delegations, or (what is not so different) to ascertain the likelypredictions, or motives, of enacting coalitions. Here courts would run upagainst not only their own institutional limitations but also the limits ofpolitical science. Notwithstanding the many plausible and instructiveexplanations for how various political arrangements have contributed toentrenchment, there is nothing like a reliable predictive model. Confrontedwith the question whether major policy reforms like the Affordable Care Act orDodd-Frank are likely to become entrenched, social scientists can do little morethan point to a number of possibly relevant variables.6' Case studies suggestthat subtle, contextual differences can often be determinative.

For all of these reasons, functional entrenchment would be considerablymore difficult to identify and police than at least some types of formalentrenchment. Even if the formal and functional varieties of entrenchment areequally harmful, therefore, legal regulation might sensibly focus only on theformer-looking under the light at formal entrenchment while leavingfunctional entrenchment in the dark.

Like many partial solutions, however, this one raises a "second-best"concern. If avenues of formal entrenchment are foreclosed, the obviousalternative would be for political actors to pursue functional entrenchmentstrategies instead. As we have emphasized, such strategies often seem to beclose substitutes for formal legislative and electoral entrenchment. No doubtthey are not perfect substitutes. Political actors who make use of formalentrenchment devices presumably do so because they are more effective or lesscostly than the functional alternatives. At the very least, however, we shouldpredict that shutting down formal entrenchment mechanisms would have theeffect of increasing the use of functional alternatives.

261. See, e.g., Patashnik & Zelizer, supra note 12, at 1079-83.

262. For example, whereas recipients of non-means-tested programs like Social Security are "atleast as active" as the public as a whole, other public-assistance programs, such as Aid toFamilies with Dependent Children (AFDC), appear to have demobilized their beneficiaries,leaving the programs quite vulnerable to retrenchment. Joe Soss, Lessons of Welfare: PolicyDesign, Political Learning, and Political Action, 93 AM. POL. Sci. REV. 363, 365 (1999) (quotingSYDNEY VERBA ET AL., VOICE AND EQUALITY: CIVIC VOLUNTARISM IN AMERICAN POLITICS 210

(1995)). Scholars explain these variable outcomes by pointing to subtle differences inprogram design, such as how the beneficiaries' relationship with AFDC caseworkers shapestheir political attitudes. See, e.g., CAMPBELL, supra note 1o8, at 129. Case studies of legislativedurability and the opposite are collected and analyzed in LIvING LEGISLATION: DURABILITY,CHANGE, AND THE POLITICS OF AMERICAN LAWMAKING (Jeffery A. Jenkins & Eric M.Patashnik eds., 2012); and PATASHNIK, supra note 12.

464

125:4oo0 2015


This substitution effect would obviously reduce the benefit of policingformal entrenchment. But it might also impose additional social costs. Unliketheir formal counterparts, functional entrenchment strategies inject a particulartype of strategic political consideration into substantive policymaking.Consider again the example of labor law. The approach to labor law thatSouthern Democrats found most effective for entrenching themselves in officemay not have been the approach to labor law they would have chosen purely asa matter of policy; at the very least, labor law would have been less of alegislative priority had it not offered an alternative path to entrenchment. Thesame is true of other political actors who have turned to labor law as a vehicleof entrenchment: but for Senator Taft's political entrenchment goals, we maynever have gotten the Taft-Hartley Act. Moreover, it seems reasonable toexpect that policy made for reasons other than entrenchment will tend to bebetter policy from the perspective of public welfare. For example, despite themany policy advantages of a carbon tax, the goal of entrenching climate changelegislation points away from such a tax and toward a cap-and-trade regime.3

In sum, functional entrenchment is doubly distorting -distorting on both apolitical and a policy margin. 6 ' This at least complicates the case for policingformal but not functional entrenchment. That approach might yield somewhatless entrenchment, but the entrenchment that does result will tend to carrygreater costs. Returning to the climate change example, we might well dobetter by permitting the formal statutory entrenchment of an efficient carbontax than by encouraging political actors to substitute a relatively inefficient cap-and-trade approach in order to duplicate the entrenchment functionally. Moregenerally, if functional entrenchment cannot be regulated effectively, policingformal entrenchment alone is not obviously the right fallback position.

The greater identifiability of formal entrenchment techniques suggests afinal reason for skepticism about making those methods the exclusive focus ofjudicial scrutiny. As generations of constitutional theorists have argued, judicialreview might be best reserved for cases of political process failure.265 To besure, entrenchment might be viewed as one such failure. But at least in somecontexts, that failure will be preventable by voters and interest groupsmounting political resistance to entrenchment efforts. The more visible theseefforts, the more resistance might be generated. If this is right, then there

263. See Lazarus, supra note 14, at 1193-95.

264. To the extent functional entrenchment strategies involve delegations, the relevant distortion

would be in the structure of government and processes of political decision making. Here,too, we should expect considerable spillover costs.

265. See, e.g., JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW

(1980).

465


might be somewhat less of a need for judicial review in the context of formalentrenchment, which is relatively visible, than in the case of functionalentrenchment, which is harder to identify and to distinguish from ordinary orbenign politics. In this regard as well, public law's current approach toregulating entrenchment may have things backwards.

3. (In) conclusion

Once the continuity between the circumscribed categories of formalentrenchment and the vast terrain of functional entrenchment has beenrecognized, could we nonetheless conclude that a sensible regulatory regimeshould continue to focus exclusively on the former while ignoring the latter?Perhaps. At least some types of legislative and electoral entrenchment are easilyidentifiable and formally distinguishable from the workings of ordinarypolitics. Even if these mechanisms of political entrenchment are no moresocially harmful than many others, the lower enforcement and error costsmight lead courts and other legal regulators to conclude that it is optimal toaddress some part of the entrenchment problem while leaving the rest alone.

We have sought to show, however, that this analysis is, at the very least,incomplete. Some types of formal entrenchment are easier to identify andquarantine than others. And the same might be true of some discrete types offunctional entrenchment, which might turn out to be no less amenable toregulatory spotlighting and excision if courts or scholars ever decided to makethe attempt. A more comprehensive assessment of the workability of policingentrenchment would have to get down to cases and institutions, in the mannerof ongoing debates over the viability of policing partisan gerrymandering andother entrenchment devices in election law. Such an assessment would alsohave to reckon with the realistic scope of the substitution and double-distortion problem: in contexts where barriers to formal entrenchment createhydraulic pressure to substitute more socially harmful strategies of functionalentrenchment, the regulatory calculus will be different. Further, the extent towhich entrenchment might be self-correcting through ordinary politicalprocesses should also bear upon the need for judicial and other forms of legalregulation. In the absence of a more comprehensive assessment along all ofthese lines -and, indeed, in the absence of any prior recognition by courts andscholars of functional entrenchment as a potential problem-there is littlereason for confidence that the prevailing approach of public law towardentrenchment is in any way optimal.

466

125:4oo0 2015


The same is true of the prior, and more fundamental, question of whetherformal entrenchment-or any other category of comparable politicalbehavior-is actually harmful or socially undesirable. As the discussion ofconstitutional entrenchment below will elaborate, it is far from obviouswhether entrenchment generally should be viewed as a democratic pathologythat we should be striving to extirpate, or, instead, as a socially beneficialconstraint on untrammeled majority rule. Probably the best answer is some ofboth. Some uses of entrenchment will be widely perceived as beneficial, othersas clearly harmful, and still others debatable. One need only consider exampleslike Social Security or Lazarus's entrenchment scheme for climate changelegislation to appreciate the difficulties and disagreements that will inevitablyarise in distinguishing socially beneficial from socially destructiveentrenchment. In these and other cases, we might wonder how the potentialbenefits of credible commitment and political stability should be weighedagainst the democratic sacrifice of present majority will, or how to distinguishthe kinds of "good" policies that should be entrenched from the "bad" onesthat should not. These difficulties and disagreements are only compoundedwhen the relevant assessment is categorical. A robust normative theory forsorting socially beneficial versus harmful types of entrenchment would have tobe combined with a well-developed empirical sense, or prediction, of thedistribution of these various kinds. But there is no reason to believe that theselines would break along a simplistic division between formal and functionalentrenchment. If public law theorists who selectively condemn electoral andlegislative entrenchment are conducting categorical cost-benefit calculationsalong these lines, they have never shown their work.

B. Extensions

1. The Uncertain Case for Distinguishing Constitutional Entrenchment

Condemnation of formal entrenchment in the electoral and legislativespheres sits uneasily with public law's longstanding, if at times ambivalent,embrace of constitutional entrenchment. The difficulty of achieving aconstitutional amendment through Article V raises well-rehearsed concernsabout the countermajoritarian, or anti-democratic, nature of constitutionalismthat run parallel to the standard set of concerns about entrenchment at theelectoral and legislative levels. Entrenched constitutions are said to substitutethe "dead hand" control of the past for present majority will. Moreover,

266. See infra Section III.B.i.

467


entrenched constitutional rules and rights can serve to lock in maladapted orevil political arrangements in much the same way as entrenchment at theelectoral or legislative level. Consider the constitutional entrenchment ofslavery, requiring a bloody civil war to eradicate, or the Supreme Court'sentrenched constitutional objections to the New Deal's recovery program,which sought to prevent the collapse of capitalism and democracy.

The accepted wisdom appears to be that constitutional entrenchmentcomes with a set of benefits outweighing these costs.6 7 But these benefits areno different in kind than the potential benefits of electoral or legislativeentrenchment. At the constitutional and subconstitutional levels alike, the risksof locking in bad decisions must be weighed against the rewards ofprecommitting to good decisions that might otherwise be sacrificed on accountof short-term interests or political pathologies -whether free speech or soundmonetary policy. Similarly, the benefits of political stability- cashed out interms of credible commitment, beneficial reliance, or dampened politicalcontestation- seem broadly similar in the constitutional and subconstitutionalcontexts.

Not surprisingly, then, some theorists have viewed constitutional andsubconstitutional entrenchment as normative equivalents -equally bad, orequally not bad. Klarman takes the former view, emphasizing thatconstitutional entrenchment shares the democratic pathologies of electoral andlegislative entrenchment, and viewing all as equally suspect."' Posner andVermeule take the opposite view, pointing to the parallels with constitutionalentrenchment as an argument for treating legislative entrenchment as no lesspernicious or permissible."

More commonly, however, public law theorists who condemn electoral andlegislative entrenchment embrace entrenchment at the constitutional level.270

In fact, the prohibition on legislative entrenchment is sometimes cast as anentrenched constitutional rule -and one that should inarguably command our

267. See, e.g., Eule, supra note 67, at 390 ("It would appear, therefore, that whatever practicalobjections can be raised against entrenchment are concerns that the Framers were willing toput aside for the benefits that barriers to change bring with them.").

268. Klarman, supra note 4, at 502-09.

269. Posner & Vermeule, supra note 24, at 1670-71 ("[A]n entrenching statute is like a mini-constitution in its self-conscious effort to control the voting practices or policy choices offuture majorities.").

270. See, e.g., Hearings, supra note 72 ("[O]nly by a constitutional amendment can one truly bindthe future: unless we keep clearly in mind that distinction between a constitutionalamendment and a bill or resolution, we have really lost our way.").

468

125:40o 2015


respect." This raises the question of what, if anything, beyond positive law

differentiates constitutional from subconstitutional entrenchment?

One possibility relates to the process through which constitutional rulesand rights are enacted. Most obviously, in contrast to ordinary legislation,constitutional ratification and amendment formally require supermajoritarianpolitical support." To the extent broader political support correlates with

better decision making, the benefits of entrenching constitutional norms mightbe higher, on average, than the benefits of entrenching the products ofmajoritarian legislative processes."' Beyond formal voting rules, a "dualist"perspective on constitutional democracy portrays constitutional norms asemerging from rare moments of "higher" politics, involving greater popularparticipation, deliberation, or public-regarding motivation than the kinds of

disengaged, compromised, and self-interested "lower" politics that ordinarilyprevail.'7 In this view, entrenching the products of higher-qualityconstitutional decision-making processes against revision through thecorrupted processes of ordinary politics might augment popular sovereigntyand improve social welfare. Entrenching the inferior products of ordinarypolitics would carry no such benefits.

Another possible basis for differentiating constitutional entrenchmentrelates to the kinds of political outcomes that are being protected againstchange. Perhaps the most important thing the Constitution has accomplishedis to establish and solidify the basic structural framework of government - thebicameral structure of Congress; the procedural outlines of the Article I,Section 7 lawmaking process; the electoral cycles and terms of office for

representatives, senators, and presidents; and the like. Entrenchment of the

basic institutional structure of political decision making may be especiallyvaluable. The alternative to settled and stable agreement at this fundamentallevel of political organization is not organized democratic contestation but

271. There is no logical contradiction here. If a prohibition on subconstitutional entrenchment

can be derived from the text and structure of the Constitution, that is reason enough to

respect it, without inferring any prejudice against the entrenchment of the Constitution

itself. Setting aside the dictates of positive law, however, the differentiating features of

constitutionalism remain to be identified.

272. Depending on the baseline being used to measure entrenchment, the symmetrical

supermajoritarian hurdles necessary for constitutional enactment and constitutional revision

could mean that constitutional law is not entrenched at all. See supra note 25 andaccompanying text.

273. See McGinnis & Rappaport, supra note 24, at 426-29 (defending the distinction betweenconstitutional and legislative entrenchment on the ground that constitutional law requires a

supermajority to enact, and supermajoritarian decision making tends to be higher quality).

274. See supra note 96 and accompanying text.

469


sheer anarchy. Moreover, the costs of entrenchment at this level may berelatively low. To the extent the constitutional structure of government createsa relatively even playing field for competing political interests- setting therules of the political game, without determining the winners and losers -fairness concerns about artificially privileging particular partisan or policyinterests will be mitigated." One could argue, then, that the constitutionalentrenchment of the basic structures and processes of political decision makingis both more valuable and less destructive than the entrenchment of specificpower holders and policy outcomes - more democratically enabling thandisabling. 6

Neither of these approaches to differentiating constitutional entrenchmentis straightforward or self-evidently persuasive. Parties, political movements,and policies that command supermajoritarian or qualitatively higher-orderdemocratic support might stake a comparable claim to entrenchment throughelectoral or legislative pathways. And electoral and legislative entrenchmentmight also contribute to locking in the basic institutional structures ofgovernment, given that most of the rules structuring the administrative state,the democratic process, the internal workings of Congress, and other aspects ofthe organization and operation of government are the product ofsubconstitutional law. Indeed, it is these two features that motivate theorists toreconceptualize some formally subconstitutional rules as functionallyconstitutional. Thus, in the view of some theorists, small-c constitutional rulesare just those that serve to "constitute" the government, a set that includes themany formally subconstitutional rules relating to the structure of governmental

75. This is clearly an oversimplification. Enumerated rights and other constitutional provisionsoperate to entrench policies in much the same way as legislative and electoral entrenchment.Thus, constitutional prohibitions on slavery and Jim Crow segregation effectively entrench apolicy regime of racial integration and equality; the First Amendment entrenches a policyslant toward religious liberty and free speech; the constitutional protection of abortionrights prevents their reversal; the Second Amendment blocks the enactment ofcomprehensive gun control laws; and so on. Constitutional structure, too, can beunderstood as a means of generating and entrenching certain policy outcomes. The originaldesign of the federal government was supposed to protect vulnerable minorities, includingcreditors, religious sects, and slaveholders, against hostile majorities, and to do so in a moredurable way than rights and other mere "parchment" barriers. Whatever is left of that idea,constitutional structure continues to generate predictable policy outcomes, or at leastgeneral biases. For example, by providing for exit and inciting jurisdictional competition,constitutional federalism makes certain kinds of redistribution more difficult to accomplishthan would be the case in a completely centralized system of government. See Weingast,supra note 163. To this extent, the policy consequences of constitutionally entrenchedfederalism will overlap in some predictable ways with the policy consequences ofentrenching Republicans in office or legislatively entrenching lower tax rates.

276. See HoLMES, supra note 28, at 6-8.

470

125:40 0 2015


institutions or the workings of the political process.' At the same time, as wehave seen, other theorists view the special democratic pedigree of statutes likethe Civil Rights Act as markers of quasi-constitutional status.' The functionallesson here is that (formal) constitutional law does not have a monopoly overthe entrenchment of structural arrangements or of the fruits of higher-orderdemocratic politics.

A functional perspective complicates the assessment of constitutionalentrenchment in other ways as well. Most fundamentally, once we look beyondthe formal notion that constitutional change can only be accomplished throughArticle V, the extent to which constitutional law is in fact politically entrenchedbecomes an open question. Constitutional rules and rights that can be changedsimply by shifting the vote of a single Supreme Court Justice (or the opinion ofa single President) may be no more entrenched, and may be quite a bit lessentrenched, than an ordinary statute." The formal vision of an entrenchedconstitution contrasts with the widely recognized reality of continuousconstitutional change through processes of judicial interpretation, politicalconstruction, and popular acceptance.8o Dramatic structural changes-such asthe growth of the administrative state, the decline of federalism, and theexpansion of presidential power -have taken place through these channels, inresponse to shifting patterns of political demand. The scope and existence ofconstitutional rights-for example, in the contexts of economic liberty, freespeech, race and gender equality, and sexual orientation- have also changedmarkedly in response to shifts in public opinion and political mobilization.81Generations of political scientists and legal scholars have documented that thecontent of constitutional law tends to converge with the preferences ofnational-level majorities."2

277. See Llewellyn, supra note 95, at 31; Young, supra note 99, at 415-16.

278. See supra notes 97-98, iol and accompanying text.

279. Recall Young's observation that the Social Security Act's promise of government financialsupport in old age is less likely to be abolished than canonical constitutional norms such asabortion rights. See supra note too and accompanying text.

28o. See generally STRAUSS, supra note 29 (arguing that the Constitution is shaped akin tocommon law to reflect contemporary norms).

281. Id.

282. See, e.g., BARRY FRIEDMAN, THE WILL OF THE PEOPLE: How PUBLIC OPINION HAS

INFLUENCED THE SUPREME COURT AND SHAPED THE MEANING OF THE CONSTITUTION

(2009); ROBERT G. McCLOsKEY, THE AMERICAN SUPREME COURT 224 (1960) (" [I]t is hardto find a single historical instance when the Court has stood firm for very long against areally clear wave of public demand."); Robert A. Dahl, Decisionmaking in a Democracy: TheSupreme Court as a National Policy-Maker, 6 J. PUB. L. 279, 285 (1957) (" [T]he policy viewsdominant on the Court are never for long out of line with the policy views dominant among

471


None of this belies the possibility that at least some aspects ofconstitutional law are at least somewhat entrenched against majoritariandemands for change. A number of the most important structural features of theU.S. government-the bicameral structure of Congress; the basic outlines ofthe Article I, Section 7 lawmaking process; the electoral cycles and terms ofoffice for representatives, senators, and presidents; and the like-haveremained mostly noncontroversial and more or less intact since the Founding,even as their original claims to functional and political efficacy have eroded.For example, it seems unlikely that Americans writing on a blank slate wouldrecreate the rule of equal state representation in the Senate; yet that rule wouldnow be formidably difficult to dislodge. Even very popular Presidents do notrun for third terms or dissolve Congress. In these and other respects,constitutional rules and rights appear to create effective constraints on politicalpreferences.

Beyond these anecdotal observations, however, we know very little aboutthe precise patterns of entrenchment that make some constitutional rules andrights more difficult to change than others.8 We also know very little aboutthe mechanisms through which functional constitutional entrenchment comesabout. We should expect, however, that these mechanisms will operate quitesimilarly in the constitutional and subconstitutional domains-by way ofselectively empowering certain groups whose interests and incentives alignwith compliance and preservation.8' The overarching point is that, whateverthe formal rules governing political and legal change, change will in fact alwaysbe possible and predictable whenever it serves the interests of powerfulpolitical actors. As James Madison famously put it, constitutional and otherlegal rules may create merely "parchment barriers" that can be ignored oroverridden at will. 8s As Madison also recognized, however, parchmentprohibitions can be converted into meaningful constraints when the politicalprocess is arranged in such a way that political actors who supportconstitutional rules and rights have the power to defend them.286 These areprecisely the kinds of political arrangements and dynamics that we saw in Part

the lawmaking majorities of the United States."); Michael J. Klarman, Bush v. GoreThrough the Lens of Constitutional History, 89 CALIF. L. REv. 1721, 1749-50 (2001) (stressinghow rarely the Supreme Court's constitutional interpretations have diverged from thepolitical preferences of national majorities).

283. See generally Levinson, supra note 17.

284. See id. at 672-716 (assimilating political and constitutional entrenchment and identifying aset of mechanisms common to both).

285. THE FEDERALIST No. 48, supra note 102, at 276.

286. See Levinson, supra note 17, at 665-70.

472

125:4oo 2015


II, operating to stack the deck in favor of subconstitutional policies and theconstituencies that support them.

Indeed, from a functional perspective, it becomes difficult to see any cleardistinction between political entrenchment at the constitutional andsubconstitutional levels. As theorists of the small-c constitution recognize, andas we have emphasized here, many subconstitutional rules and arrangementsare at least as entrenched as many constitutional norms (in addition to sharingother indicia of constitutionality, like heightened democratic support orstructural significance). And many formally constitutional rules and rightsseem not to be especially entrenched against functional change. Rather thanviewing constitutional entrenchment as a separate and superior category,public law scholars might do better to recognize the continuity of politicalentrenchment at the constitutional and subconstitutional levels.

2. The Uncertain Categorical Boundaries ofEntrenchment

Up to now, we have attempted to show that the kinds of politicalarrangements and consequences that are described as "entrenchment" in theformal electoral, legislative, and constitutional contexts have functionalanalogues that share all of the relevant features and that cannot bemeaningfully distinguished. In advancing these arguments, we have taken forgranted that political entrenchment can be usefully understood as a distinctiveand reasonably well-defined phenomenon. But once we move beyond formaldefinitions of entrenchment, the boundaries of the category, and hence itsanalytical value, begin to blur.

As we have seen, entrenchment tends to be identified with barriers put inplace by upstream decision makers that impede political changes preferred bypresent majorities or that make the status quo more difficult to change than itwas to create in the first place. Upon reflection, however, there areinnumerable political arrangements that fit these descriptions.

As a first cut, to the extent entrenchment is identified with impediments topresent majority will, it becomes difficult to distinguish from the many othercountermajoritarian features of the U.S. political system. Some of thesefeatures are a product of constitutional design.8' In addition to the formalsupermajorities required to ratify a treaty or convict an impeachment,bicameralism and separation of powers, combined with the different electoral

287. On the countermajoritarian and more generally counterdemocratic features imposed by theU.S. Constitution, see ROBERT A. DAHL, How DEMOCRATIC IS THE AMERICANCONSTITUTION? (2001); and SANFORD LEVINSON, OUR UNDEMOCRATIC CONSTITUTION

(2006).

473


bases of the Senate, the House, and the President, impose a de factosupermajority requirement for enacting legislation. The constitutionalmalapportionment of the Senate and the Electoral College, combined withinternal legislative rules and arrangements, like the Senate filibuster and thecongressional committee system, create further minority veto gates that areoften used to block majority-preferred actions. And of course, constitutionalrights and judicial review at least sometimes impose countermajoritarianconstraints. Beyond the constitutional structure of government, numerousother institutional arrangements and familiar features of democratic politicsconspire against perfect responsiveness to majoritarian preferences: thedisproportionate influence of well-organized, well-connected, or wealthyminorities and interest groups; the vagaries of cycling and agenda setting;information deficiencies; political geography; and so on.

In short, frustration of majority will-and even supermajority will-is apervasive feature of American politics.88 Against this background, demarcatinga discrete category of entrenchment defined in terms of political constraints onpresent majority will presents a conceptual challenge. It is not obvious on whatbasis the paradigm cases of formal electoral and legislative entrenchment or thefunctional analogues we presented in Part II should be distinguished from thebroader universe of countermajoritarian features of the American democratic'system.99

Another intuitive understanding of entrenchment emphasizes theasymmetrical difficulty of reversing prior political decisions and, as aconsequence, the disproportionate influence of past decision makers overpresent and future ones. Thus, political entrenchment is thought to arise when"a temporary political majority (in the society and in the legislature) . . .extend[s] its hold on power into the future,"290 or when a political action"limits the policy choices available to future governments."9 '

Here again, however, it is hard to see the boundaries of any categorydefined in this way. Nearly every political action -as well as every instance ofinaction -has some constraining effect on the choices of downstream decision


289. One limiting feature of entrenchment, as compared to the broader category of

countermajoritarianism, is that entrenchment is focused on impediments to political change.Yet countermajoritarian political decision making can also be a source of political change.Minorities can block political changes preferred by majorities, but they can also enactpolitical changes dispreferred by majorities.

29o. Klarman, supra note 4, at 498.

291. Serkin, supra note 3, at 888.

474

12S:4oo 2015


makers." At a minimum, every political decision will generate different "factson the ground" that subsequent political actors will have to reckon with.Political actors at Time One who decide (not) to start a war, slow globalwarming, foster economic growth, borrow money, or invest in education willbequeath different states of the world to their successors-and thereforedifferent political options, with different costs, benefits, and distributiveconsequences. Decisions at Time Two will invariably depend on theenvironment and political calculus created by decisions at Time One.

One way of limiting the category of entrenchment is to focus on thepurpose or motivation of present decision makers. As we have seen, concernsabout formal, legal entrenchment have focused on intentional efforts by partiesand politicians who are self-consciously seeking to secure their hold on poweror the durability of their policies at the expense of rivals.293 (As we have alsoseen, functional entrenchment strategies can be deployed intentionally andstrategically in just the same way.) But if the salient feature of entrenchment iscontrol of the future by the past, it is not obvious why the purposes ormotivations of political decision makers should matter at all. From theperspective of the present, the constraints imposed by past decisions will be thesame regardless of whether they were intended or entirely inadvertent.

As it happens, the phenomenon of entrenchment as it has been understoodby social scientists and historians (and even the occasional legal scholar) is inno way limited to the self-consciously strategic efforts of political actors. To thecontrary, in many contexts the focus is on "entrenched" rules, policies, andinstitutions that have become socially and politically difficult to change owingto path-dependent processes of political development that were never intendedor foreseen by their creators.294 In fact, social scientists view this kind ofunintentional entrenchment, or "lock in," as a quite general phenomenon,resulting from a somewhat predictable set of political dynamics -including, inparticular, the political feedback effects of empowering winners whiledisempowering losers that we highlighted in Part II."s

a2. See Dana & Koniak, supra note 67, at 530-31 (recognizing that "in some sense all acts ofpresent legislature [s] -for example, decisions about whether to declare war, how muchmoney to print, whether to invest in infrastructure, and whether to invest in education-limit what future legislatures might do"); Klarmai, supra note 4, at 504-05 ("[V]irtually anyaction taken by today's majority may (deleteriously) affect the future."); Serkin, supra note

3, at 888 ("In principle, this definition is sufficiently broad to encompass every single actthat a government undertakes.").

293. See supra Section I.B.

294. See Starr, supra note lo.

295. See id. at 23-26.

475


Examples can be found in nearly every area of policy and politics. Incontrast to President Roosevelt's strategic posture toward entrenching SocialSecurity, there is no reason to believe that the designers of Medicare wereplotting entrenchment, yet the program appears just as deeply entrenched as aresult of more or less the same set of political dynamics. 9 Or consider thehome mortgage interest deduction, which created- apparently quite byaccident -a constituency of homeowners and mortgage lenders that is deeplycommitted to, and formidably capable of, preserving their entitlement.97

Along the same lines, the casual, short-sighted decision of Congress to exemptexisting coal-fired plants from the stringent regulations of the 1970 Clean AirAct gave rise to a powerful interest group that served as a major impediment tosubsequent antipollution and climate change measures.298 Legal scholars havedocumented how rules of corporate law relating to ownership structure havebecome entrenched through a political-economic feedback cycle in whichexisting rules and arrangements increase the wealth and political power ofcorporate stakeholders who have an interest in maintaining or enhancing theserules and arrangements.299 Regulatory regimes governing corporate ownershipstructure and telecommunications have become increasingly difficult to modifyas influential stakeholder groups become more deeply invested in, and betterable to defend, existing arrangements.3

oo

Relatively localized examples like these could be multiplied countlessly, butinadvertent, functional entrenchment also operates on a broader scale. Forinstance, the political economy of modern capitalism probably involves asimilar "rich get richer" dynamic, in which economic winners wield their newlyacquired wealth and power to preserve and augment institutions that allowthem to become ever more wealthy and politically influential.30 ' Even morebroadly, Acemoglu and Robinson's account of why some nations becomewealthy while others fail describes a similarly self-reinforcing historical

296. See id. at 31-35.

297. See Hal R. Varian, An Opportunity To Consider if Homeowners Get Too Many Breaks, N.Y.TIMES (Nov. 17, 2005), http://www.nytimes.com/2oo5/1n/17/business/17scene.html [http://perma.cc/9WXV-JFGC].

298. See Richard L. Revesz, "War on Coal" or "Original Sin"? Power Plants, Public Health, andClimate Change (Sept. 1, 2014) (unpublished manuscript) (on file with authors).

299. See Lucian Arye Bebchuk & Mark J. Roe, A Theory ofPath Dependence in Corporate Ownershipand Governance, 52 STAN. L. REV. 127, 157-60 (1999).

300. See id. (discussing corporate ownership structure); Starr, supra note to, at 27-29 (discussing"policy-technological lock" in the context of communications policy).

301. See, e.g., LARRY M. BARTELS, UNEQUAL DEMOCRACY: THE POLITICAL ECONOMY OF THE NEW

GILDED AGE (2008) (describing how this political-econornic dynamic has worked in thecontemporary United States).

476

125:4oo0 2015


process, through which an initial set of political arrangements produceseconomic consequences that in turn reinforce the initial political arrangements,resulting in deeply entrenched political-economic regimes that persist forcenturies -well beyond the time horizons or design capabilities of any group ofpolitical actors.3o2 Along the same lines, any kind of path-dependent account ofhistorical processes will portray initial choices or contingencies as shapingfuture decisions and outcomes in a manner that could be assimilated toentrenchment. For example, the vast bodies of work exploring the "historicallegacies" of colonialism, communism, or slavery, documenting broad patternsof economic, political, and social development that stem from prior conditions,could be understood as addressing entrenchment writ large.303

Recognizing the general path-dependence of human history is obviously along leap from the specific phenomena of legislative and electoralentrenchment and their close functional analogues. But that is the point. If thesalient feature of entrenchment is that past political decisions continue to holdsway over the present, then entrenchment is a vastly broader category than iscommonly perceived. To the extent history matters - which is to say, to avirtually unlimited extent-present political decisions will be shaped andconstrained by prior choices. Ongoing rule by the dead hand of the past is aubiquitous and unavoidable feature of temporally extended democracy.

This is not to prejudge any effort toward usefully differentiating a morecircumscribed category of political entrenchment. It is merely to point out theabsence of conceptual resources for doing so in the existing literature onentrenchment. Once we blur the distinction between formal and functionalentrenchment, the question of how else the category of political entrenchmentmight be usefully bounded becomes open.

CONCLUSION

Courts and legal scholars have focused considerable attention on theproblem of political entrenchment. Yet their focus has been oddly myopic,

302. See supra note ii and accompanying text. Other accounts of the development of the systemof modern capitalism-both of its origins starting in sixteenth-century Europe and of itsculmination in contemporary America- describe a similar process of self-reinforcingentrenchment, in which the rich get richer and more powerful and use their snowballingadvantages to entrench capitalist arrangements. See Daron Acemoglu et al., The Rise ofEurope: Atlantic Trade, Institutional Change, and Economic Growth, 95 AM. ECON. REV. 546(2005).

303. See Starr, supra note lo, at 36-45.

477


narrowly directed toward the formal "rules of the game"3o4 governing elections,legislation, and constitutional amendment, each considered as its ownindependent domain. By failing to appreciate both the common denominatorsamong the various impediments to political change in these areas and the vastterrain of functional analogues, public law has avoided confrontingentrenchment as a general feature of the democratic process. This narrow focushas limited the extent to which courts and scholars have developed a coherentunderstanding of what political entrenchment is, why or when it is bad, andwhat-if anything-legal regulators should do about it.

This Article has sought to broaden the frame that public law places aroundentrenchment. Its core contribution has been to call attention to the ways inwhich politicians, parties, and policies can be entrenched against change notjust by changing the formal rules of the political game but by playing the gamein strategic ways: enacting substantive policies that strengthen politicalallies or weaken political opponents, shifting the composition of the politicalcommunity, or altering the structure of government decision making. Thesekinds of functional entrenchment strategies allow political actors to achieve thesame results without resorting to the kinds of formal rule changes that wouldraise red flags from the perspective of public law. Recognizing the continuity offormal and functional entrenchment thus invites the question of why publiclaw identifies and condemns the former while ignoring or pardoning the latter.

It also raises a broader set of questions about the phenomenon of politicalentrenchment. To start, the familiar set of commitment and stability benefitsstemming from entrenchment in many different contexts, ranging fromsovereign debt to environmental stewardship, complicates the common (if notconsistent) public law intuition that impediments to political change should besuspect. Political entrenchment, whether formal or functional, has potentialbenefits as well as costs that must be assessed contextually and empirically.Public law's longstanding embrace of entrenchment by way ofconstitutionalism is a rather significant illustration of this point. What is it thatdifferentiates constitutional entrenchment, or the entrenchment of monetarypolicy through independent central banks, from the kinds of electoral andlegislative entrenchment that public law regards as obviously problematic?Under what conditions should we view the entrenchment of politicians,parties, or policies as beneficial, benign, or pathological? And then, at aconceptual level, once we move beyond the myopic focus on formalentrenchment and expand our frame of vision to encompass close functionalsubstitutes, the boundaries of the category of political entrenchment become

304. Issacharoff& Pildes, supra note 26, at 709.

478

125:40 0 2015


difficult to discern. How should we understand and differentiate entrenchmentfrom the inevitable influence of the past on the future and the ubiquitous path-dependence of political decision making? Definitive answers to these questionsawait further inquiry, but an important first step is to frame the questionsclearly and explain why they are important, as we have attempted to do here.

The analysis of entrenchment presented in this Article feeds into severalmore general currents in public law and democratic theory. There is a vastliterature in political science and law on the various mechanisms of democraticdesign that impede or facilitate political change. These include studies ofsupermajority and other kinds of voting rules that impose varying degrees ofdifficulty on departures from the status quo; "temporary" legislation and othertiming rules that formally require or expire political actions;os institutionalarrangements that create or mitigate gridlock in political decision-makingprocesses;3o6 the propelling and paralyzing effects of presidential versusparliamentary systems, or party-unified versus divided government;30 7 the roleof the judiciary in accelerating or braking legal and political transformations;"o0critical theories advocating greater "plasticity" in democratic and socialstructures;30 9 and, of course, constitutionalism.31 o Viewing political

entrenchment alongside these and other theoretical perspectives on the paceand pathways of political change might be mutually illuminating.

Our analysis of entrenchment also links to other areas of public law inwhich functional perspectives have complicated conventional understandingsof formal rules and arrangements. Some of these connections we have noted:the reality of constitutional change through channels other than the Article Vamendment process,"' the dependence of constitutional entrenchment andconstraint upon underlying political commitments to maintain and respect

305. See FRANK FAGAN, LAW AND THE LIMITS OF GOVERNMENT: TEMPORARY VERSUS PERMANENT

LEGISLATION (2013); Jacob E. Gersen, Temporary Legislation, 74 U. CHI. L. REV. 247 (2007);Jacob E. Gersen & Eric A. Posner, Timing Rules and Legal Institutions, 121 HARv. L. REV. 543(2007).

306. See, e.g., KEITH KREHBIEL, PIVOTAL POLITICS: A THEORY OF U.S. LAWMAKING (1998).

307. See Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 HARv. L.REV. 2312 (2006).

308. Compare Michael J. Klarman, Brown, Racial Change, and the Civil Rights Movement, 8o VA. L.REV. 7 (1994), with Mark Tushnet, The Significance of Brown v. Board of Education, 8o VA.

L. REV. 173 (1994).

309. See ROBERTO MANGABEIRA UNGER, PLASTICITY INTO POWER: COMPARATIVE-HISTORICAL

STUDIES ON THE INSTITUTIONAL CONDITIONS OF ECONOMIC AND MILITARY SUCCESS (1987).

310. See, e.g., MELISSA SCHWARTZBERG, DEMOCRACY AND LEGAL CHANGE 156-59 (2007) (arguingthat constitutional entrenchment is at odds with fundamental democratic values).


479


constitutional rules and rights,' and the existence of "superstatutes" and otherformally subconstitutional policy instruments that share some (if not all) of thefunctional attributes of constitutionality."' Other recent works in public lawdescribe how party politics have functionally transformed the formal systemof separation of powers314 and federalism,"'s how the formally counter-majoritarian capacity of judicial review has been tempered by the functionalrealities of politics,31' how states and localities exercise meaningful forms ofpower despite the erosion of their formal sovereignty,17 and how the politicalreality of presidential leadership and executive dominance has come to trumpthe legal formalities of a more limited executive role."' Many more examplescould be added to this list, but the general point should be clear: politicalentrenchment is far from the only area in which the formal legal rulesdo not fully capture-and may in fact obscure-fuller and more realisticunderstandings of how political processes and institutions function forpurposes of public law.

Finally, our account of political actors' ability to use functionalentrenchment devices to navigate around prohibitions on formal entrenchmentseems of a piece with a broader body of scholarship cataloguing the waysmotivated parties avoid the dictates of public law by resorting to legallypermissible means of accomplishing ends that public law intends to prohibit.Such public law "workarounds" take many forms, potential and actual.319 Forexample, although the Constitution contemplates a President elected accordingto the votes of the Electoral College, states could ensure that the President waselected by a national popular majority by directing their electors to vote for the

312. See supra notes 93, 278-281 and accompanying text.


314. See Levinson & Pildes, supra note 307.

315. See Jessica Bulman-Pozen, Partisan Federalism, 127 HARV. L. REV. 1077 (2014).

316. See FRIEDMAN, supra note 282.

317. See Heather K. Gerken, Of Sovereigns and Servants, 115 YALE L.J. 2633 (20o6); Benjamin I.Sachs, Despite Preemption: Making Labor Law in Cities and States, 124 HARV. L. REV. 1153(2011).

318. See ERIC A. POSNER & ADRIAN VERMEULE, THE EXECUTIVE UNBOUND: AFTER THEMADISONIAN REPUBLIC (2010).

319. See Mark Tushnet, Constitutional Workarounds, 87 TEx. L. REV. 1499 (20o9). Tushnet saysthat "constitutional workarounds" are possible "(a) when there is significant politicalpressure to accomplish some goal, but (b) some parts of the Constitution's text seem fairlyclear in prohibiting people from reaching that goal directly, yet (c) there appear to be otherways of reaching the goal that fit comfortably within the Constitution." Id. at 1503.

480

125:4oo0 2015


person who wins the national popular vote. 0 Even the Constitution's corerights-granting provisions are subject to workarounds. For example, although

abortion cannot be banned directly, the procedure can be put out of reachthrough the imposition of multiple confining regulations known as TRAPs(targeted regulation of abortion providers).' More generally, notwithstanding

the unconstitutional conditions doctrine, governments are often permitted to

accomplish indirectly what they are forbidden from doing directly by using theincentive of conditional funding.' In the statutory context, campaign financelaw's susceptibility to workarounds is well known and well captured by themetaphor of "hydraulics."' When the law closes one channel for political

spending, political actors simply open up substitute channels. Similarly, stateand local actors can often find ways to avoid the preemptive effect of federalstatutes.? For example, although federal law preempts essentially all state andlocal regulation of union organizing, states and cities have averted this barthrough creative partnerships with unions and employers." Similarpossibilities for working around environmental, ERISA, and immigrationpreemption are also available. * Thus, functional entrenchment is anotherreminder that political actors can often navigate around public law rules, and

that public law often serves to channel means rather than dictate ends.

320. See id. at 1500. For another example, the infamous "Saxbe fix" has been used to work aroundthe Emoluments Clause, allowing, for instance, Hillary Clinton to be appointed Secretary ofState despite the fact that the salary for that position had been increased while Clinton wasserving in the Senate. See id. at 1501.

321. Cf Virginia: Targeted Regulation of Abortion Providers (TRAP), NARAL PRo-CHOICE Am., http://www.prochoiceamerica.org/government-and-you/state-governments/state-profiles/virginia.html [http://perma.cc/T6VR-XLNH] (describing various statutoryregulations that effectively limit access to abortion).

322. For example, constitutional constraints mean that the government cannot directly prohibitprivate organizations from offering abortion counseling services, but the federal governmentwas able to move in this direction by conditioning the grant of funding for family planningservices on recipients' agreement not to offer abortion counseling. See Rust v. Sullivan, SooU.S. 173 (1991). Similarly, constitutional limitations made it at least uncertain whether thefederal government could mandate that states set their drinking age at twenty-one, butCongress was able to work around any such constitutional restriction by withholdingfederal highway money from states that allowed people under twenty-one to drink. SeeSouth Dakota v. Dole, 483 U.S. 203, 206 (1987).

323. See Samuel Issacharoff & Pamela S. Karlan, The Hydraulics of Campaign Finance Reform, 77TEX. L. REv. 1705 (1999).

324. See Sachs, supra note 317, at 1157-60.

325. See id. at 1154-56.

326. See id. at 1222.

481


We leave these further pathways of investigation for another day. For now,our more limited ambition has been to show that the conceptual boundariesand normative implications of political entrenchment overflow the formal andnarrowly categorical terms in which the phenomenon has been cast in publiclaw. At the very least, courts and scholars, along with voters and citizens,would do well to recognize that it is not only the arcane possibility of legislativeentrenchment or the transparently strategic features of election law likegerrymandering and voter ID requirements that should be viewed through theprism of entrenchment. Labor law reform ought to be understood as a partisanbattle between incumbent Democrats and Republicans fighting over their ownelection prospects, and not only as a contest over union organizing rights andlabor-management relations. The same is true for immigration law, tort law,and many other policy initiatives. It is also true for the creation of centralbanks, human rights treaties, and the question of whether or not the District ofColumbia should become a state. And it is true about environmental policy andthe Affordable Care Act. Recognizing that these and many other policies andpolitical arrangements have important implications for the prospects andpathways of change opens a new perspective on entrenchment as a pervasivefeature of democratic politics.

482

125:4oo0 2015

I~I

Documents

Political Entrenchment and Public Law